The Challenge of a Global Standard of Justice: Peace, Pluralism, and Punishment at the International Criminal Court

Published on May 2016 | Categories: Types, Business/Law | Downloads: 46 | Comments: 0 | Views: 275
of 74
Download PDF   Embed   Report

Abstract: This article addresses what is often described as the "peace versus justice" problem, in the context of prosecutorial discretion at the International Criminal Court. The problem typically arises when the threat of prosecution would derail peace negotiations or deter a tyrant from relinquishing power. It already confronts the ICC in its first referral, concerning crimes against humanity committed during an on-going civil war in Uganda, where many victims and their families are imploring the prosecutor to foreswear prosecution. They argue that indictments will deter the rebels from negotiations, and that their traditional restorative justice mechanisms will serve to promote justice and reconciliation far better than prosecutions. This article analyzes the conflicting claims of peace, pluralism and punishment in such cases. Among the central questions explored in the article are the following: (1) Does justice in the aftermath of crime always require prosecution? The answer explores the nature of retributive justice and its relationship to victims of crime. (2) If justice does require prosecution, does this obligation outweigh all other considerations? The analysis considers three alternative ways of handling a conflict between the obligations of justice and the dire impact fulfilling them would have on third parties, when such is the case. (3) As a global institution, how much deference should the ICC afford to diverse state approaches to the previous two questions? The article suggests that for procedural, substantive, and pragmatic reasons, the Court and its prosecutor should adopt a pluralist philosophy in its charging decisions and complementarity assessments. ICC decisions on these issues will shape the contours of an emerging standard of global criminal justice.

Comments

Content

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

The Challenge of a Global Standard of Justice: Peace, Pluralism, and Punishment at the International Criminal Court
ERIC BLUMENSON* This Article addresses what is often described as the “peace versus justice” problem, as it confronts the recently-established International Criminal Court (ICC). The problem typically arises when the threat of prosecution would derail peace negotiations or deter a tyrant from relinquishing power. If a state then grants amnesty or de facto impunity as the price of peace, should the ICC’s prosecutor bring charges in its stead? This Article analyzes the conflicting claims of peace, pluralism, and punishment in such cases by exploring three fundamental questions: (1) Does justice in the aftermath of crime always require prosecution and punishment? (2) If justice does require prosecution, does this obligation outweigh all other considerations? (3) As a global institution, how much deference should the ICC afford to diverse state approaches to the previous two questions? Given the Rome Statute’s silence on these questions, the ICC will have to develop its own answers, and these answers will help shape the contours of an emerging global standard of criminal justice.
∗ Professor of Law, Suffolk University Law School. J.D., Harvard Law School, 1972. I am grateful to the International Criminal Court and Chief Prosecutor Luis Moreno Ocampo for the opportunity to serve in the Office of the Prosecutor as a visiting legal scholar in 2004, and to the attorneys and investigators in that office who offered me an indispensable education in the work and promise of the ICC. Thanks are also due to Jeff Alsdorf, Tom Clark, Stan Fisher, Simon Keller, Gabriel Levin, Eduardo Rivera Lopez, Katherine Lu, David Lyons, Kate Martin, Eva Nilsen, Guido Pincione, Pat Shin, Ken Simons, Marion Smiley, Carol Steiker, and Ellen Winner, for their comments and counsel on this project; to Peter Shorett, for expert research assistance throughout; and to participants in law faculty colloquia at Boston College, Boston University, Suffolk University, Universidad Torcuato Di Tella, and the University of Arizona, at which I presented earlier versions of this Article.

Electronic copy available at: http://ssrn.com/abstract=834004

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

802

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

INTRODUCTION .................................................................................802 I. THE PROBLEM ASSERTED: OPPOSITION TO THE ICC’S FIRST CASE .......................................................................................806 II. MORAL STAKES IN THE DECISION TO PROSECUTE ..................818 A. Three Grounds for Prosecution and Punishment ...........818 1. The Institutional Responsibility Claim.....................820 2. The Consequentialist Claim .....................................823 3. The Retributive Justice Claim ..................................829 B. When Peace and Prosecution Conflict...........................832 1. Does Justice Always Demand Prosecution?.............834 2. Do Obligations of Justice Override All Else? ..........841 a. The Deontological Purist’s View: Clean Hands Above All ...............................................844 b. The Minimize Injustice View ............................846 c. The Catastrophic Exception View: “Threshold Deontology”....................................849 C. Pluralism ........................................................................853 III. THE MORAL ACCEPTABILITY OF CONDITIONAL AMNESTIES .............................................................................859 A. Vindicating the Victims .................................................861 B. Holding Perpetrators Accountable.................................868 CONCLUSION ....................................................................................871

INTRODUCTION This Article concerns an issue that will compel the International Criminal Court (ICC)1 to resolve some fundamental controversies about its proper role and potential contributions. The issue is whether the ICC should ever defer to a state’s amnesty of its present or past leaders for heinous crimes, and if so, when. Such amnesties take many morally distinguishable forms. Chile’s Pinochet decreed his own amnesty, but the people of Uruguay voted
1. The ICC became effective on July 1, 2002, with the ratification of the Rome Statute by sixty countries. Rome Statute of the International Criminal Court, July 17, 1998, U.N. Doc. A/CONF.183/9, 37 I.L.M. 1002, 2187 U.N.T.S. 90 [hereinafter Rome Statute]. The Statute establishes ICC jurisdiction over war crimes, genocide, crimes against humanity, and (if and when defined) aggression, if committed by the national of, or on the territory of, a state party or a state which has accepted jurisdiction by declaration. Id. arts. 5, 12. However, under its complementarity principle, such a case is inadmissible if it has been genuinely investigated or prosecuted by a state with jurisdiction, or if it is of insufficient gravity. Id, art. 17. The ICC sits in The Hague, Netherlands.

Electronic copy available at: http://ssrn.com/abstract=834004

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

803

to grant amnesty to their former leaders, and the post-apartheid South African government granted amnesties only to individual applicants who provided a full confession. Sometimes an amnesty has been declared to achieve legitimate goals the country deems of overriding importance, such as peace, truth, or reconciliation; sometimes its sole purpose is impunity. In this Article, I argue for the moral acceptability of amnesties-for-truth under conditions that, I believe, can further the ICC’s mission of ending the culture of impunity; and I argue that the ICC is in a position to enunciate and encourage these conditions. Amnesties have haunted the human rights era that began with the Nuremberg trials. Tyrants typically have been unwilling to relinquish power without a guarantee of safe passage or amnesty, and insurgents often have been unwilling to negotiate surrender without some kind of immunity provision.2 They have prevailed often enough that the former United Nations High Commissioner for Human Rights was probably correct in stating that “[a] person stands a better chance of being tried and judged for killing one human being than for killing 100,000.”3 Despite large advances in the last decade,4 these dynamics favoring criminal impunity remain, and at least until its promise is realized, the ICC will undoubtedly confront situations in which the cost of prosecution may be continued tyranny or bloodshed. What should happen in such cases is highly controversial, and so politically fraught and complex that the Rome Treaty negotiators were unable to agree to any provision on the issue.5 Instead, they left
2. See Jack Snyder & Leslie Vinjamuri, Trials and Errors: Principle and Pragmatism in Strategies of International Justice, 28 INT’L SECURITY 5, 5 (2003). 3. Statement by José Ayala Lasso, Overview, Establishment of an International Criminal Court, http://www.un.org/law/icc/general/overview.htm (last visited Mar. 8, 2006); see also Ruth Wedgwood, The International Criminal Court: An American View, 10 EUR. J. INT’L L. 93, 95 (1999) (“[D]emocratic transitions in this century have generally occurred in circumstances that did not permit the criminal trial of outgoing leaders.”); Michael P. Scharf, The Amnesty Exception to the Jurisdiction of the International Criminal Court, 32 CORNELL INT’L L.J. 507, 508–09 (1999) (“Case studies in Haiti, South Africa, and the Dayton Accords demonstrate that the offer of amnesty may be a necessary bargaining chip to induce human rights violators to agree to peace and relinquish power.”). 4. In addition to the creation of the three international tribunals—for Yugoslavia, for Rwanda, and the ICC—and hybrid tribunals combining international and national elements—such as tribunals in Sierra Leone, Kosovo, Cambodia, and East Timor—many countries have revisited crimes of the past and attempted to replace impunity with belated accountability. See, e.g., Larry Rohter, After Decades, Nations Focus on Rights Abuses, N.Y. TIMES, Sept. 1, 2005, at A4; Paul van Zyl & Mark Freeman, Conference Report, in THE LEGACY OF ABUSE: CONFRONTING THE PAST, FACING THE FUTURE 3, 6–8, 13–14 (Alice H. Henkin ed., 2002). 5. Mohamed M. El Zeidy, The Principle of Complementarity: A New Machinery to

Electronic copy available at: http://ssrn.com/abstract=834004

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

804

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

it to the Prosecutor and judges of the ICC to develop an approach to amnesty and truth commissions over time.6 As it happens, that time may be now, because in the first referral to the prosecutor—massive atrocities committed in an on-going civil war in Northern Uganda— many in the victim community are urging the ICC to forgo prosecution in favor of a traditional Ugandan reconciliation process called mato oput.7 They claim prosecutions would violate their traditions, and also prolong the war by making it too costly for the rebels to negotiate or surrender.8 To address such claims, one must range widely. The amnesty issue is not an isolated policy question but one that compels the Office of the Prosecutor to grapple with issues concerning its appropriate role and responsibilities, and the proper weights it should attach to the claims of peace, pluralism, and punishment when they conflict. While legal and pragmatic constraints will play an important role, these decisions also require fundamental moral judgments on three inescapable and extraordinarily difficult issues: (1) A question of justice: Does justice in the aftermath of crime always require prosecution? In transitional situations, states have sometimes tried to impose accountability through truth commissions, reparations, traditional confession and reintegration rituals, or other non-penal means. If some such methods impose a sufficient degree of accountability and justice, it may be possible for the ICC to avoid a forced choice between peace without justice and justice without peace. (2) A question of impact: If justice does require prosecution, does this obligation outweigh all other considerations? May the prosecutor reject a prosecution on the ground that it threatens harm to innocent third parties, or is he obligated to bring criminal charges against serious offenders despite grave political or human costs? (3) A question of pluralism: As a global institution, how much deference should the ICC afford to diverse state approaches to the previous two questions? These questions are in part philosophical but, given the stakes, hardly academic. They exist at the heart of the ICC, and embody what I characterize, in the title of this Article, as the challenge of a global standard of justice. By this I do not mean
Implement International Criminal Law, 23 MICH. J. INT’L L. 869, 941–42 (2002). The Rome Treaty established the ICC and promulgated its governing law, the Rome Statute. 6. Id. 7. See infra notes 27–35. 8. See infra notes 23–35 and accompanying text.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

805

merely that the ICC is a global organization that must devise its own jurisprudence. More importantly, if it works as envisioned, the ICC will set national standards as well, defining what domestic criminal justice should look like for adjudicating crimes against humanity, genocide, and war crimes. This is a consequence of the Rome Statute’s principle of complementarity, which allows the ICC to proceed with a case only when a state is not genuinely doing so itself.9 Under this provision, the Court will have to issue explicit decisions about the adequacy of a particular state’s response to these crimes. Over time, ICC decisions will define minimum requirements for criminal justice systems throughout the world—establish, in other words, a global standard of criminal justice. In what follows, I begin by describing and disentangling the weighty issues that surround the amnesty question. Part I offers a short narrative of the situation extant in the prosecutor’s first referral, concerning crimes against humanity committed in Uganda. It then distinguishes several additional kinds of amnesty scenarios, placing them in a decision matrix that helps reveal the different legal and moral postures presented by each case. Part II analyzes the moral considerations that should inform the prosecutor’s exercise of discretion in these various situations. It examines the grounds for privileging justice, peace, and pluralism separately, and argues that although amnesia in the aftermath of extreme crimes can never be justified, in some circumstances individualized amnesties in exchange for truth may comport with the diverse values underpinning each of these three concerns. Part III considers one example, the South African Truth and Reconciliation Commission (TRC), and suggests that it should be considered a morally acceptable response to extreme criminality under all of the preceding arguments. Although I have adopted an argumentative approach, a principal aim in what follows is diagnostic, not prescriptive. The Article endeavors to disentangle the distinct strands that come together in the demand for “justice above all,” and detail some essential considerations regarding each that should inform the decisions of the ICC and its prosecutor. Although I argue for a capacious embrace of state-centered, sometimes non-penal accountability mechanisms, there are no knockdown arguments on the issue, just arguments that draw support more from their
9. Rome Statute, supra note 1, art. 17. For purposes of Article 17, a state is not investigating or prosecuting when it either fails to pursue the case, or pursues it in a way that demonstrates its inability or unwillingness to genuinely investigate or prosecute the case. Id. art. 17; see infra note 43 and accompanying text.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

806

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

verisimilitude than from logical proofs. But although my proposed policy is contestable, I believe the series of questions I consider to get there are not. They are questions that must be addressed by proponents on all sides; perhaps doing so would uncover common ground where assertions of deeply-held convictions do not. I. THE PROBLEM ASSERTED: OPPOSITION TO THE ICC’S FIRST CASE

Two of the three referrals accepted by the chief prosecutor so far involve on-going wars, a situation that multiplies the stakes and complexity of his decisions.10 As an example, consider the first case referred to the ICC: atrocities inflicted on the Acholi people of Northern Uganda in the course of a war between government troops and a rebel force known as the Lord’s Resistance Army (LRA).11 This war, now nineteen years running, has resulted in tens of thousands of civilian deaths and the displacement of over 1.5 million people—approximately ninety percent of the population in Northern Uganda.12 Apart from its leaders, the LRA is largely a children’s army. Approximately eighty-five percent of the soldiers are children, most between the ages of eleven and fifteen, who were kidnapped and impressed into the army.13 After the initial abduction, which in many cases includes forcing the child to murder or mutilate family members in order to destroy the child’s bonds with his community, the LRA puts its
10. At this point, the ICC has accepted three state referrals (from Uganda, the Democratic Republic of Congo (DRC), and the Central African Republic, respectively) and one referral from the U.N. Security Counsel regarding crimes in the Darfur region of Sudan. See Situations and Cases, International Criminal Court, http://www.icc-cpi.int/cases.html (last visited Mar. 9, 2006). The prosecutor’s decision-making framework is discussed infra at notes 36–44 and accompanying text. 11. See Decision Assigning the Situation in Uganda to Pre-Trial Chamber II, ICC02/04 (July 5, 2004), available at http://www.icc-cpi.int/cases/current_situations/Uganda/ ug_decision.html (follow “English” hyperlink below “Decision assigning the situation in Uganda to Pre-Trial Chamber II”). 12. For a general discussion of the LRA, see Meera Selva, The Mystic and his Brutal Army of Child Soldiers, THE INDEPENDENT (London), July 30, 2004, at 32, 33. 13. President of Uganda Refers Situation Concerning Lord’s Resistance Army (LRA) to International Criminal Court, M2 PRESSWIRE, Jan. 30, 2004 [hereinafter President of Uganda]. For a discussion of the life conditions of LRA child soldiers, see Angela Veale & Aki Stavrou, Violence, Reconciliation and Identity: The Reintegration of Lord’s Resistance Army Child Abductees in Northern Uganda 23–26 (Institute for Security Studies, Monograph Series No. 92, 2003), available at http://www.issafrica.org/Pubs/Monographs/ No92/Contents.html.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

807

children through elaborate deprivation and initiation rituals, and trains them in the sparse spiritual principles of its leader Joseph Kony, a self-styled New Testament prophet.14 Beatings are frequent and insubordination can be a capital offense, according to exsoldiers.15 Amidst all this, these child soldiers are ordered to loot and burn houses, kidnap other children, and kill both civilians and fellow soldiers.16 To avoid this rampaging army, approximately 30,000 children walk for hours from their villages to Gulu every night, where they can sleep in tents in relative safety, returning the next morning.17 Northern Uganda has become a land full of dispossessed people, many of whom have had their lips, breasts, or ears sliced off as a demonstration of rebel power.18 Apart from its long-running war effort, the Ugandan government has periodically attempted to offer incentives to the LRA rebels to lay down their arms. There have been several attempts at peace negotiations.19 More significant thus far have been successive amnesty laws intended to promote surrender and reconciliation, the broadest of which was enacted in 2000. The law grants immunity against criminal liability to Ugandan rebels in exchange for surrender, with no requirement that the applicant describe the crimes he committed or participate in any restorative justice mechanism.20
14. Kony’s spiritual doctrine is reported as little more than an obligation to destroy all forms of evil and promote the Ten Commandments, although he also claims to receive detailed military guidance from holy spirits. See Melanie Thernstrom, Charlotte, Grace, Janet and Caroline Come Home, N.Y. TIMES, May. 8, 2005, at 34. 15. As one child ex-soldier reported, “Whoever tries to escape will be killed, for walking ahead, you will be killed, and even for a minor mistake children will be severely tortured.” Veale & Stavrou, supra note 13, at 25. 16. President of Uganda, supra note 13. 17. Payam Akhavan, Developments At The International Criminal Court: The Lord’s Resistance Army Case: Uganda’s Submission Of The First State Referral To The International Criminal Court, 99 AM. J. INT’L L. 403, 409 (2005). By 2003, the population living or sleeping in the region’s refugee camps numbered 800,000 people, or seventy-five percent of the region’s population. Id. 18. Christopher Ringwald, A War on Children: Rebels with no Agenda Terrorize Uganda’s, NAT’L CATHOLIC REP., May 13, 2005, at 13, available at http://ncronline.org/ NCR_Online/archives2/2005b/051305/051305a.php. A recent survey in Northern Uganda showed that twenty-three percent of respondents had been mutilated during the conflict, and forty percent had previously been abducted by the LRA. See PHUONG PHAM ET AL., FORGOTTEN VOICES: A POPULATION-BASED SURVEY ON ATTITUDES ABOUT PEACE AND JUSTICE IN NORTHERN UGANDA 4 (2005), available at http://www.ictj.org/downloads/ ForgottenVoices.pdf. The percentages in the Acholi region are surely much higher, because this survey of 2500 people interviewed in Spring 2005 included four northern districts—two mostly Acholi and two mostly non-Acholi. Id. at 3. 19. Ringwald, supra note 18, at 13 (reporting on peace talks held in the early 1990s and 2004). 20. Amnesty Act, pt. II (2000) (Uganda), available at http:www.c-r.org/accord/

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

808

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

It also establishes agencies to assist in the resettlement and reintegration of former LRA soldiers.21 As of early 2005, approximately 14,000 soldiers had fled the LRA and several other rebel groups to seek amnesty.22 But the amnesty had not succeeded in ending the war or obtaining the surrender of the LRA leadership or most of its troops. The most recent government strategy is seemingly the opposite of amnesty: seeking prosecution of the LRA leadership before the ICC.23 But almost immediately following the referral, a
Uganda/accord11/keytext.shtml (follow “The Amnesty Act, 2000” hyperlink). Under this law, an amnesty is available to Ugandan citizens only, for all crimes they committed “in the cause of” war against the government after January 26, 1986. The Amnesty Act has been subsequently renewed many times and remains in effect today. Ugandan Gov’t Extends Ceasefire with Northern Rebels, XINHUA NEWS AGENCY, Dec. 17, 2004, § World News; Uganda: “Speed Up Peace Talks,” ALL AFR. NEWS (THE MONITOR), Dec. 15, 2004, available at http://allafrica.com/stories/200412150176.html (last visited Mar. 27, 2006). The crimes covered include genocide, murder, kidnapping, and rape (which a previous amnesty did not), apparently in response to strong popular sentiment. Earlier amnesties had restricted amnesties to less serious crimes, but no such limitations exist in the more recent versions because they were “not supported by the Ugandan communities and a ground-swell of opinion pushed for a blanket amnesty based on principles of reconciliation.” Veale & Stavrou, supra note 13, at 34; see also Barney Afako, Reconciliation and Justice: “Mato Oput” and the Amnesty Act, 11 ACCORD: PROTRACTED CONFLICT, ELUSIVE PEACE (2002), available at http://www.c-r.org/accord/uganda/accord11/reconcilliation.shtml (stating that Ugandans and their and their religious leaders “rejected the partial proposals and strongly advocated the adoption of a general amnesty”). 21. Amnesty Act, supra note 20, §§ 11, 13. 22. Over 14,000 Ugandan Rebels Get Amnesty, XINHUA NEWS AGENCY, Mar. 2, 2005, § World News; Uganda: “Speed Up Peace Talks,” supra note 20. These numbers vary somewhat from report to report. The government stated in its 2003 report to the Human Rights Committee that amnesty had brought the surrender and return of 5000 former LRA soldiers. U.N. Human Rights Comm., Consideration of Reports Submitted by State Parties Under Article 40 of the Covenant, ¶¶ 128, 131(b), U.N. Doc. CCPR/C/UGA/2003/1 (Feb. 25, 2003). 23. Press Release, Int’l Criminal Court, President of Uganda Refers Situation Concerning the Lord’s Resistance Army (LRA) to the ICC (Jan. 29, 2004), http://www.icccpi.int/press/pressreleases/16.html. ICC prosecutor Luis Moreno Ocampo then negotiated the terms of the referral to include crimes committed by the Ugandan military as well. Interview by Martha Minow with Luis Moreno Ocampo, Chief Prosecutor, Int’l Criminal Court, in Cambridge, Mass (Jan. 19, 2005) (author in attendance) [hereinafter OcampoMinow interview]. Ugandan President Museveni promised the ICC that he would seek to amend the amnesty laws to exclude top LRA leaders. Id. Nevertheless, Museveni has continued to equivocate on the issue of blanket amnesty. In September 2004, Museveni said that “if it would help heal the wounds in Acholi, the government would consider petitioning the ICC that the LRA leadership not be charged criminally, but be allowed to go through the traditional ceremony of ‘mato oput’ (cleansing).” Uganda: Kony Can Talk to Bigombe—Museveni, ALL AFR. NEWS (NEW VISION), Sept. 7, 2004, available at http://allafrica.com/stories/200409071661. html. But as Amnesty International points out: [A]s soon as the situation has been referred, the ICC has jurisdiction and the

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

809

number of interest groups and government factions announced their opposition to ICC prosecution, arguing that it would short circuit the amnesty and reconciliation process and prolong the war.24 These Acholi leaders argue two points relevant here. First, they claim that replacing the amnesty offer with ICC prosecution would undermine peace negotiations with LRA leaders by removing their incentive to give up arms and surrender. One prominent member of the Acholi Religious Leaders’ Peace Initiative (ARLPI) said, “[n]obody can convince a rebel leader to come to the negotiating table and at the same time tell him that when the war ends he will be brought to trial.”25 LRA representatives themselves recently claimed that they want to negotiate a settlement, but not if they would face ICC prosecution.26 Second, opponents argue that ICC prosecution holds less promise than the Acholi’s traditional conflict resolution method, mato oput, which stresses forgiveness, reconciliation, and reintegration of offenders. Barney Afako, a Ugandan human rights lawyer and consultant to the Amnesty Commission, says that among the Acholis, “many offences [sic], including homicides, have traditionally been resolved by reconciliation. Whenever a homicide takes place the rwodi [traditional chiefs] intervene in the situation to ‘cool down the temperature’ and to offer mediation.”27 In the mato
state cannot “withdraw” its referral. Under Article 86 of the Rome Statute it then has the absolute duty to “cooperate fully with the Court in its investigation and prosecution of crimes within the jurisdiction of the Court.” Press Release, Amnesty International, Government Cannot Prevent International Criminal Court from Investigating Crimes (Nov. 16, 2004), http://news.amnesty.org/index/ ENGAFR590082004. 24. In early 2004, the head of the Uganda government’s Amnesty Commission joined members of Parliament and religious leaders from Northern Uganda in opposing the ICC investigation. Kony: Amnesty, not Arrest, ALL AFR. NEWS (THE EAST AFRICAN), Feb. 16, 2004, available at http://allafrica.com/stories/200402170874.html. 25. Carlos Rodriguez, Situational Report Northern Uganda (Jan. 27, 2004), http://www.austria-uganda.at/_report_northern_uganda.htm. Additionally, the Acholi opponents claim that even if the amnesty were redrawn to exclude only the LRA’s leaders, it would still effectively deter many rank-and-file soldiers from abandoning rebel forces and returning to their communities. See Whose Justice? Perceptions of Uganda’s Amnesty Act 2000: The Potential for Conflict Resolution and Long Term Reconciliation 23 (Refugee Law Project, Paper No. 15, 2005), available at http://www.refugeelawproject.org [hereinafter Refugee Law Project Paper 15]. 26. Interview by Matthew Brubacher, Office of the Prosecutor of the ICC, with Author, in The Hague, Neth. (Mar. 10, 2005). After a decade without contact, the LRA and the government began negotiations in 2004—“the best opportunity for peace that northern Uganda has had since the war began,” according to the International Crisis Group. Katherine Southwick, Op-Ed., When Peace and Justice Clash, INT’L HERALD TRIB., Oct. 14, 2005, at 6. 27. Afako, supra note 20.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

810

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

oput process, opposing parties come before a group of elders, who in turn identify the cause of the conflict through questioning.28 After the guilty party accepts its responsibility and demonstrates repentance, “[t]he elders lay down terms of compensation and reconciliation is sealed by sharing a bitter root drink from a common calabash . . . .”29 Although there is a serious question as to the efficacy of this process in fostering either truth or reconciliation, especially when applied to mass criminality, the Acholi response has been to institutionalize reforms to the process rather than turn to criminal justice.30 The Paramount Chief of Acholi, Rwot Onen David Acana II, opposes trying Kony at the ICC. “I wonder who will help them in giving evidence to prosecute Kony since the Acholi do not buy their idea of taking him to court . . . ,” said Acana. “[T]he Acholi have forgiven all LRA rebels and their leader Kony of crimes committed against them . . . .”31 It appears that a substantial majority of the Acholi people, who comprise both the victims and the perpetrators of the war with the LRA, want reconciliation and favor extending amnesty for all rebels. A survey conducted by Uganda’s Refugee Law Project reported majority support for the amnesty and antagonism towards ICC intervention among the victim community.32 (Part of this, of course, is that they want their children home, rather than kidnapped,
28. Creating Space for Traditional Peace-Building Mechanisms in Africa, LUTHERAN WORLD INFO, Oct. 23, 2002, available at http://www.lutheranworld.org/News/LWI/EN/ 1088.EN.html [hereinafter Creating Space]. 29. Press Release, African Rights, Northern Uganda: Justice in Conflict (Jan. 20, 2000), available at www.africanrights.unimondo.org/html/ugand001.html. In cases of weapon-related violence, spears are exchanged between the conflicting parties and bent at the tips. “From then on, all the parties must vow not to harm one another, since they are united by the ritual and therefore rendered brothers and sisters.” Creating Space, supra note 28. 30. Zarchary Olum, chairman of the Acholi Group within the Ugandan Parliament, argues that “[a]mnesty should be extended without amendment. Consideration should be given to local initiatives for reconciliation. The cultural approach, mato-oput, should be emphasized . . . .” Don’t Push ICC on Kony—Acholi, ALL AFR. NEWS (NEW VISION), Aug. 19, 2004, available at http://allafrica.com/stories/200408190191.html; see also Refugee Law Project Paper 15, supra note 25, at 25–27; Workshop Report, Uganda Human Rights Comm’n, The Implications of the International Criminal Court Investigations on Human Rights and the Peace Process in Uganda 12–13 (Eunice Apio & Isaac Bakayana rapportuers, Oct. 5, 2004) (on file with author) [hereinafter Workshop Report] (reporting Uganda human rights conference view that although mato out system has loopholes it is preferable to any other processes because it facilitates reconciliation while holding perpetrators responsible). 31. Acholi Chief Opposes Kony Trial, ALL AFR. NEWS (NEW VISION), Nov. 8, 2004, available at http://allafrica.com/stories/200411081457.html. 32. Refugee Law Project Paper 15, supra note 25, at 1, 4, 9, 21, 28. The 2000 amnesty was enacted after a government fact-finding mission found widespread demand for it. Workshop Report, supra note 30, at 11.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

811

jailed, or shot.33) However, as of this writing, the Chief Prosecutor has just obtained arrest warrants for leaders of the LRA, against the advice of the Ugandan mediator who had arranged last year’s peace talks between the two sides.34 Prosecutor Luis Moreno Ocampo believes that pursuing the case does not undermine the possibilities of peace negotiations or surrender in this conflict, although he appears open to the mato oput process for those bearing less responsibility.35 Distinguishing Cases and Their Legal Status In the first instance, it is the Office of the Prosecutor that decides how to respond to the Ugandan situation and others that present similar concerns. The Rome Statute affords the prosecutor proprio motu power, which authorizes him to initiate an investigation or prosecution, although both decisions are subject to judicial review by the pretrial chamber and to deferral by the UN Security Council.36 The prosecutor was assigned this power over the objection of many delegations who wanted the ICC’s caseload limited to situations referred by state parties, the Security Council, or both.37 By
33. Workshop Report, supra note 30, at 9. A more recent and broader survey, extending beyond the Acholi region, found that although sixty-six percent of respondents believed offenders should be prosecuted, that figure fell to twenty-nine percent if amnesty were the only road to peace. PHUONG PHAM ET AL., supra note 18, at 4–5. 34. Uganda Aide Criticizes Court Over Warrants, N.Y. TIMES, Oct. 9, 2005, at 20 (The mediator, Betty Bigombe, stated that “[t]here is now no hope of getting them to surrender . . . . I have told the court that they have rushed too much”). 35. See Ocampo-Minow interview, supra note 23; see also Press Release, Int’l Criminal Court, Joint Statement by ICC Prosecutor and the Visiting Delegation of Lango, Acholi, Iteso and Madi Community Leaders from Northern Uganda (Apr. 16, 2005), available at www.icc-cpi.int/press/pressreleases/102.html (stating that all parties will “integrate the dialogue for peace, the ICC and traditional justice and reconciliation processes”); Darryl Robinson, Serving the Interests of Justice: Amnesties, Truth Commissions and the International Criminal Court, 14 EUR. J. INT’L L. 481, 500 (2003) (stating that prosecuting only those most responsible, and allowing others to receive amnesty for testimony, “could pass the complementarity test, assuming that the number of persons selected for prosecution were not so small that it is evidently a gesture to avoid real prosecutions”). 36. See Rome Statute, supra note 1, arts. 53 1(c), 2(c), 3(b) (pretrial chamber review, which includes the authority of the pretrial chamber to review sua sponte a decision not to proceed); id. art. 16 (Security Council deferral). For discussion on judicial oversight of the prosecutor’s decisions, see Allison Marston Danner, Navigating Law and Politics: The Prosecutor of the International Criminal Court and the Independent Counsel, 55 STAN. L. REV. 1633, 1646–48 (2003). 37. Silvia A. Fernandez de Gurmendi, The Role of the International Prosecutor, in THE INTERNATIONAL CRIMINAL COURT: THE MAKING OF THE ROME STATUTE 175, 180–82 (Roy S. Lee ed., 1999); Jenia Iontcheva Turner, Nationalizing International Criminal Law, 41 STAN. J. INT’L L. 1, 5–7 nn.19–28 (2004). In the years since the establishment of the Court, the

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

812

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

providing for direct prosecutorial referrals, the nature of the ICC was transformed from a court that would be tethered to state interests and big-power vetoes, to a court that in theory may pursue egregious situations wherever they lead. Article 53 of the Rome Statute also authorizes the prosecutor to decline to investigate or prosecute an otherwise admissible case if doing so would not serve the interests of justice.38 In the next section, I explore in detail the moral considerations that are relevant to the prosecutor’s decision whether to pursue prosecutions. Preliminarily, let us put the Ugandan example in a broader context, to better assess the legal framework within which the prosecutor operates. We can distinguish six relevant scenarios by means of the following matrix:

United States has highlighted the proprio motu power as a central argument in its broadside against the Court. The Bush administration argues that the Rome Statute created a powerful yet wholly unaccountable prosecutor who has every opportunity to execute a political agenda. It claims that this agenda will be directed against the United States, whose peacekeeping efforts send soldiers all over the world and garner resentment from international organs. Consequently, the Bush Administration has engaged in a continuous effort to remove its citizens and leaders from the prosecutor’s reach, including, inter alia, “unsigning” the Rome Statute, withdrawing American aid from countries that have refused to agree not to deliver U.S. citizens to the ICC, and even signing a law that authorizes the government to use “all means necessary” to free any U.S. soldier held at the ICC—a law Europeans sometimes refer to as the “Hague Invasion Act.” See Int’l Criminal Court American Servicemembers’ Protection, 22 U.S.C.S. § 7427 (2005); Juan Forero, Bush’s Aid Cuts on Court Issue Roil Neighbors, N.Y. TIMES, Aug. 19, 2005, at A1; Coalition of the Int’l Crim. Ct., Status of U.S. Bilateral Immunity Agreements, http://www.iccnow.org/?mod=bia (last visited Apr. 20, 2006); Eric Schwartz, Op-Ed., The U.S. Assault on World Criminal Court, BOSTON GLOBE, July 1, 2002, at A11; Danner, supra note 36, at 1635–36, 1654. Even if the first instance authority of the prosecutor were as broad as the United States claims, it would have little relation to his actual power because the prosecution’s authority to initiate a case is not matched by the power necessary to pursue it. Whether he can do so depends on ratification by other organs of the Court, and, given the prosecution’s negligible investigative and arrest powers, often on the cooperation of the involved state. See infra notes 51–52. 38. Rome Statute, supra note 1, art. 53.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

813

The vertical, upward axis identifies two degrees of “fallout”—negative consequences—that would result from ICC intervention: none and dire. The horizontal, rightward axis marks three degrees of accountability imposed by the state, in increasing order: none, non-penal accountability (for example, via a truth commission that names names), and state prosecution. Different cells in the matrix present distinct moral and legal situations for the prosecutor. The rightmost column encompasses situations legally beyond the reach of the ICC prosecutor. Under the Rome Statute’s complementarity regime, when a genuine state prosecution has occurred, the case is inadmissible before the ICC. Conversely, cell four, combining state impunity with ICC efficacy, presents the clearest case for prosecution on this chart. The hard cases to be explored here are represented by cells one, two, and five. Cases in cell one confront the prosecutor with a “peace vs. justice” dilemma: doing nothing results in impunity because of the state’s blanket amnesty, but bringing charges threatens disaster for innocent third parties. It may be possible to choose a third option by delaying action until a more propitious time. If not, a prosecutor who wishes to decline an admissible referral based on its deleterious consequences would have to invoke Article 53 of the Rome Statute, which permits the prosecutor to forgo an investigation or prosecution if doing so would not serve the interests of justice.39 Whether the
39. Different provisions govern a decision to forego investigation and a decision to

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

814

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

projected destructive impact of a prosecution actually fits within the rubric of “interests of justice” will likely generate substantial debate.40 Article 53 does not provide any definition, only a nonexclusive list of factors that must be included in the balance (including, significantly, the interests of victims).41 Cases in cell five raise the issue of when, if ever, the ICC should defer to a state’s non-penal method of imposing accountability. Increasingly, States have invoked a number of such mechanisms, including truth commissions, reparations, traditional confession and reintegration rituals, and ineligibility for government
decline prosecution. Regarding the former, see Rome Statute, supra note 1, art. 53. Paragraph 1(c) provides that the prosecutor must consider whether “taking into account the gravity of the crime and the interests of victims, there are nonetheless substantial reasons to believe that an investigation would not serve the interests of justice.” Id. art. 53, para. 1(c). Regarding the latter, Article 5, paragraph 2(c) permits non-prosecution when it is “not in the interests of justice, taking into account all the circumstances, including the gravity of the crime, the interests of victims and the age or infirmity of the alleged perpetrator, and his or her role in the alleged crime.” Id. art. 5, para. 2(c). In both cases, Article 53, paragraph 3 generally allows the Pre-trial Chamber to review a decision of the prosecutor not to proceed on interests-of-justice grounds, and requires it to do so on request of the referring state or, in the case of a Security Council referral, the Security Council. Id. art. 53, para. 3. 40. What considerations can be included as “interests of justice” is, of course, highly controversial. As Kent Greenawalt notes, “the line between justice and ordinary conditions of social welfare can be thin.” Kent Greenawalt, Amnesties Justice, in TRUTH V. JUSTICE 189, 205 n.13 (Robert I. Rotberg & Dennis Thompson eds., 2000). Chief Prosecutor Luis Moreno Ocampo posed the same question, quoting a Cambodian as saying that in his country, “justice . . . is a cup of rice.” Chief Prosecutor Luis Moreno Ocampo, Lecture at Harvard Law School (Jan. 19, 2005) (on file with author). In the West, justice has historically been taken to be something quite different, and much narrower, than social welfare, but the International Covenant on Economic, Social and Cultural Rights (ICESCR) treats many social welfare issues as human rights and thus aspects of justice. Avril McDonald and Roelof Haveman note a further ambiguity, regarding whose justice is contemplated by this provision—anyone affected by a prosecution, or only the victims and defendants associated with the crime? See Avril McDonald & Roelof Haveman, Expert Consultation Process on General Issues Relevant to the ICC Office of the Prosecutor: Prosecutorial Discretion—Some Thoughts on “Objectifying” the Exercise of Prosecutorial Discretion by the Prosecutor of the ICC 6 (Apr. 15, 2002), available at http://www.icccpi.int/library/organs/otp/mcdonald_haveman.pdf; Matthew R. Brubacher, Prosecutorial Discretion Within the International Criminal Court: Finding the Balance between Law and Politics, 2 J. INT’L CRIM. JUST. 71, 79–80 (2004) (describing factors in interests-of-justice test). 41. The enumerated factors are the gravity of the crime, the interests of victims, and, in the case of a decision whether to prosecute, the age or infirmity of the alleged perpetrator, and his or her role in the alleged crime. Rome Statute, supra note 1, art. 53, paras. 1(c), 2(c). Impact on third parties is neither included nor excluded. On this issue, see Claudia Angermaier, The ICC and Amnesty: Can the Court Accommodate a Model of Restorative Justice, 1 EYES ON THE ICC 131 (2004) (noting that although the circumstances listed are very crime-specific, the fact that Article 53 directs the prosecutor to take all circumstances into account supports the view that it may consider political contingencies in its assessment).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

815

employment. A prosecutor who believes that a state’s particular alternative procedure warrants deference would most likely invoke the interest-of-justice provision here as well.42 But there is also a possible argument that the complementarity principle applies to render the case inadmissible.43 Cell two presents both issues, which may not be uncommon. By most accounts, it represents South Africa’s turn to the TRC as a version of traditional justice that could avert bloodshed and promote reconciliation between groups that had previously existed in the hostile relationship of oppressors and oppressed.44 As described by the Acholi activists, it also represents the Ugandan situation, but the

42. The argument would be that an alternative and democratically established process is underway that would better serve the interests of justice than a criminal prosecution in the Hague. Again, there will be controversy concerning the definition of justice, in this context particularly including the question whether criminal justice must be penal and retributive to count as justice. See Robinson, supra note 35, at 483, 488 (stating that “there may be exceptional circumstances where it would not be in the interests of justice to interfere with a reconciliation mechanism,” and that the statutory provision, which includes compassionate considerations, is not confined to the interests of retributive justice, but includes a broader considerations of justice). But see Christopher Keith Hall, Expert Consultation Process on General Issues Relevant to the ICC Office of the Prosecutor: Suggestions Concerning International Criminal Court Prosecutorial Policy and Strategy and External Relations 28–29 (2003), available at http://www.icc-cpi.int/library/organs/otp/hall.pdf (arguing that the interests-of-justice provision should never be applied to amnesties that prevent a judicial determination of guilt). 43. Under Article 17, a case is inadmissible if a state has investigated but decided not to prosecute, “unless the decision resulted from the unwillingness or inability of the State genuinely to prosecute.” Rome Statute, supra note 1, art. 17(b). Some have suggested that an argument could be fashioned that would describe a South African-type truth commission process as an investigation that terminated in a decision to grant amnesty, not out of any “purpose of shielding the person concerned from criminal responsibility” (a definition of unwillingness in the Rome Statute, supra note 1, arts. 17(2)(a), 2(c)), but as a necessary means of achieving peace and reconciliation while imposing public accountability. Robinson, supra note 35, at 501. It is a difficult argument, but “at least conceivable” according to a prosecuting official charged with interpreting the complementarity principle. Id.; see also Richard J. Goldstone & Nicole Fritz, “In the Interests of Justice” and Independent Referral: The ICC Prosecutor’s Unprecedented Powers, 13 LEIDEN J. INT’L L. 655, 661 (2000) (stating that South African-type amnesty processes “do not entail an absence of investigation”). But see Claudia Angermaier, The ICC and Amnesty: Can the Court Accommodate a Model of Restorative Justice, 1 EYES ON THE ICC 131, 144 (2004) (“[I]t appears that investigations before a truth commission . . . cannot render a case inadmissible before the ICC.”). 44. See, e.g., Paul Van Zyl, Dilemmas of Transitional Justice: The Case of South Africa’s Truth and Reconciliation Commission, 52 J. INT’L AFF. 647, 650 (1999); ARYEH NEIER, The Trouble with Amnesty, in WAR CRIMES: BRUTALITY, GENOCIDE, TERROR, AND THE STRUGGLE FOR JUSTICE 96, 104 (1998); Jennifer Llewellyn, Justice for South Africa: Restorative Justice and the Truth and Reconciliation Commission, in MORAL ISSUES IN GLOBAL PERSPECTIVE 96, 100–01 (Christine Koggel ed., 1999).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

816

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

claim is highly contested on both counts.45 In any particular case, the ICC and its prosecutor must respond to whichever scenario it confronts; but over time, through its decisions, the ICC itself will influence which of the scenarios come to exist. That, of course, is the very idea of deterrence, and its aspirational ideal is that all of these scenarios will wither away because crimes against humanity will become increasingly uncommon. This chart highlights more incremental possibilities. For example, a consistent prosecutorial policy that defers to state non-penal accountability mechanisms but ignores political dynamics might induce states to utilize truth commissions rather than blanket amnesties when criminal prosecution would threaten disaster.46 I turn now to the moral considerations that should inform the prosecutor’s exercise of discretion in situations that present serious consequentialist and pluralist concerns. Two assumptions will facilitate this inquiry. First, I shall assume that the prosecutor has broad legal discretion to decide whether to pursue a case using his proprio motu power, and whether to decline a referral on interest of justice grounds (although subject to review by the Court and severe practical constraints).47 There are, however, significant legal questions regarding the extent of the prosecutor’s discretion to decline to prosecute a grave crime within ICC jurisdiction. Beyond
45. Some find the traditional justice methods insufficient to impose any sort of accountability, and others deny that ICC prosecution poses any significant threat to what they regard as the remote prospects of a peace agreement. Chief Prosecutor Luis Moreno Ocampo’s public statements regarding the Ugandan situation suggest that he regards its traditional justice mechanism as a mode of alternative accountability, but believes ICC prosecution of LRA leaders poses little threat to Northern Uganda’s peace prospects. See infra notes 51–52. 46. For example, in Uganda, Acholi parliamentarians have drafted an addendum to the ICC bill, the implementing law, to attach penalties to their traditional justice mechanism in an effort to fall within the complementarity principle and prevent criminal prosecution of such cases. Christian Aid, Dialogue Session on the Implications of the Amnesty Amendments and ICC Investigations 5, 6 (Kampala, July 28, 2004) (remarks of Amnesty Group and Barney Afako) (on file with the author). 47. Discussions of “prosecutorial discretion” are complicated by the multiple meanings encompassed by the term. It may be used to describe (1) the range of decisions open to the prosecutor in the first instance; (2) whether these decisions are subject to reversal by judicial or other higher authorities, which in this case they are, see supra notes 37, 40; and finally, (3) the degree to which the prosecutorial discretion afforded in theory is realistically available in practice—and in this case one central challenge for the Office of the Prosecutor is that its freedom of action, while quite expansive on paper, is severely limited in practice by its insufficient resources and its lack of police and process powers. The latter means that, absent Security Council involvement, the prosecutor will usually have to depend on the State involved to deliver the defendants and provide access to witnesses—cooperation which is highly unlikely if state agents are themselves the targets.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

817

the issue of how narrowly one should construe the “interests of justice” provision, there is a question whether other treaties and customary laws that impose a duty on states to prosecute certain crimes also limit the discretion of the ICC’s prosecutor to decline a legally meritorious and admissible case.48 I avoid those questions here, satisfied that the prosecutor’s first instance discretion is broad enough so that an exploration of the moral considerations that should inform that discretion is not wasted effort. 49 I will also assume that the amnesties at issue extend to the most responsible and culpable perpetrators. To press the amnesty issue, we must address the hardest case, one involving amnesty of the central and most culpable figures involved—a Pinochet, a Botha, or a
48. The Rome Statute clearly authorizes and anticipates prosecutorial selectivity. See infra note 52. However, the scope of the prosecutor’s discretion in amnesty cases implicates two legal issues in particular: (1) Does the ICC’s prosecutor have the right to prosecute someone who has received a domestic amnesty? There should be little question that he does. See, e.g., Decision on the Extradition of Ricardo Miguel Cavallo, Supreme Court of Mexico (June 10, 2003), 42 I.L.M. 888 (2003); Prosecutor v. Kallon & Kamara, Case Nos. SCSL-2004-15-AR72(E), SCSL-2004-16-AR72(E), Decision on Challenge to Jurisdiction: Lomé Accord Amnesty, Special Court for Sierra Leone, ¶¶ 67, 87–89 (SCSL App. Ch., Mar. 13, 2004) (domestic amnesty cannot bar prosecution by others of crimes subject to universal jurisdiction); Prosecutor v. Furundzija, Case No. IT-95-17/1-T, Judgment, ¶ 155 (Dec. 10, 1998) (jus cogens prohibition on torture); Leila Nadya Sadat, Exile, Amnesty and International Law 66 (Wash. U. Sch. of L., Working Paper No. 05-04-03, 2005) (“[C]ases to date have unanimously concluded that the amnesties cannot ‘travel’ with efficacy to other jurisdictions.”), available at http://ssrn.com/abstract=712621; Diba Majzub, Peace or Justice? Amnesties and the International Criminal Court, 3 MELB. J. INTL’L L. 247 (2002). (2) How broad is the prosecutor’s legal right to decline to prosecute an individual granted amnesty for a grave crime within ICC jurisdiction? The answer depends in part on how narrowly one should construe the “interests of justice” requirement in Article 54 of the Rome Statute. It also depends on a construction of treaties and customary law prohibiting genocide, war crimes, torture, and terrorism, which either impose a duty on states to prosecute certain crimes or afford victims a right to a remedy. The argument that such duties extend to the ICC as well is strengthened by Article 21(3) of the Rome Statute, supra note 1, which requires the Court to apply and interpret the statute’s provisions “consistent with internationally recognized human rights.” For discussion on this question, see McDonald & Haveman, supra note 40, at 7; Majzub, supra, at 247; Scharf, supra note 3, at 507; Robinson, supra note 35, at 490–92; Ronald C. Slye, The Legitimacy of Amnesties Under International Law and General Principles of Anglo-American Law: Is a Legitimate Amnesty Possible?, 43 VA. J. INT’L L. 173 (2002); Naomi Roht-Arriaza, Amnesty and the International Criminal Court, in INTERNATIONAL CRIME, PEACE, AND HUMAN RIGHTS: THE ROLE OF THE INTERNATIONAL CRIMINAL COURT 77 (Dinah Shelton ed., 2000). 49. I will also assume that the decision confronting the prosecutor is whether to prosecute a case, although similar issues may arise regarding whether to investigate a referral. The Rome Statute’s interests-of-justice test for declining to pursue a case is arguably more liberal at the prosecution stage than the investigation stage, given the variation in factors listed. Compare Rome Statute, supra note 1, art. 53(1)(c), with id. art. 53(2)(c).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

818

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

Milosevic—who many commentators argue should never be afforded amnesty, even if contingent on confession or cooperation.50 I note that given its limited resources and powers, less responsible offenders will not be brought before the ICC in any event,51 and the prosecutor has implied that non-criminal state or tribal mechanisms may be acceptable in those cases.52 It is increasingly common for states to adopt a mixed system involving prosecution of the leaders and traditional or restorative mechanisms for the others, particularly among states which would be overwhelmed otherwise.53 The most passionate and difficult controversy concerns those leaders most responsible for atrocities and crimes against humanity. II. A. MORAL STAKES IN THE DECISION TO PROSECUTE Three Grounds for Prosecution and Punishment

Given the Prosecutor’s discretion and the necessity of selective rather than universal prosecutions, what choices should he make when confronting claims that a prosecution would produce
50. See, e.g., Neil J. Kritz, Coming to Terms with Atrocities: A Review of Accountability Mechanisms for Mass Violations of Human Rights, 59 LAW & CONTEMP. PROBS. 127, 129 (1996); Stephen Landsman, Alternative Responses to Serious Human Rights Abuses: Of Prosecution and Truth Commissions, 59 LAW & CONTEMP. PROBS. 81, 90 (1996); Hall, supra note 42, at 28. 51. ICC prosecutions will be extremely selective for three reasons. First, the Rome Statute makes cases of “insufficient gravity” inadmissible. Rome Statute, supra note 1, art. 17. Second, the ICC lacks the resources to bring to justice the vast numbers of criminals usually involved in the crimes within its jurisdiction. Luis Moreno Ocampo, the Chief Prosecutor, has said that he expects to prosecute no more than five or six cases in the next five years. Ocampo-Minow interview, supra note 23. It appears that the OTP will focus its efforts on leaders who bear the greatest responsibility, leaving most crimes within the Court’s jurisdiction to be prosecuted by national authorities or not at all. See Office of the Prosecutor, Paper on some Policy Issues, §§ 2.1, 2.2 (on file with author) [hereinafter Paper on some Policy Issues]. Finally, some of the most consequential and egregious cases will remain out of reach because of the ICC’s extremely limited enforcement powers. The Pinochets and Pol Pots of the future will generally be immune from ICC prosecution while they hold power, because absent a Security Council referral, the Court is dependent on the state’s cooperation to arrest defendants. Rome Statute, supra note 1, art. 59. 52. Paper on some Policy Issues, supra note 51, § 2.2. 53. The extreme example is playing out in Rwanda: Of the 125,000 suspects arrested between 1994 and 1998, only 5000 have been tried by a Rwandan judiciary that was itself largely destroyed by the genocide, and 80,000 remain in prison. See Kritz, supra note 50, at 135. When it has concluded its work, the International Criminal Tribunal for Rwanda (ICTR) will have tried only an infinitesimal percentage of suspects. Consequently, the Rwandan authorities have divided defendants into four categories based on the gravity of the alleged offenses, and reconstituted its traditional village-based dispute resolution process, gacaca, to handle the vast majority of cases. See infra note 211.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

819

more harm than good? To the degree possible, these decisions should be governed by principled criteria reflecting the purposes for which the ICC was created and the contributions it can best achieve. When they are, the prosecutor is exercising reasoned discretion rather than acting in arbitrary or politically biased ways. Crimes against humanity and genocide, typically committed by means of mass murders, dismemberments, kidnappings, and gang rapes, are subject to universal jurisdiction because they are crimes against all humankind.54 To allow Pol Pot, Pinochet, and others guilty of such crimes to go unpunished is a form of legal amnesia that appears to excuse the most egregious deeds, betray the victims who endured them, and encourage similar crimes against others. Consequently, there are many who believe that the ICC should adhere to a strict agenda of bringing such criminals to justice, without allowing their doomsday threats, even if credible, to derail this mission.55 Someone might argue for this policy on any of several bases. One is the consequentialist ground that crimes of this magnitude must be punished in order to prevent their recurrence through the deterrent, incapacitative, or norm-reinforcing effects of punishment. Second is the retributive justice claim that criminals must be punished as a moral imperative, not because of any social benefit that will result but simply because they deserve it. Many retributivist supporters of the ICC believe that bringing war criminals to justice is an absolute moral obligation of the Court, or a non-negotiable right of the victims. Finally, one might derive a strict duty to prosecute from a conception of the ICC’s institutional purpose: War criminals must be prosecuted, not because this is a requirement of justice simpliciter, but because it is a requirement that applies specifically to the ICC and its prosecutor as a matter of their institutional roles. In the remainder of this section, I offer some comments regarding the premises of each of these three arguments before undertaking a detailed exploration of the retributive claim, the one arguably absolutist objection to non-prosecution, in Part II.B.
54. See RESTATEMENT (THIRD) FOREIGN RELATIONS LAW OF THE UNITED STATES § 404 cmt. a (1987); Filartiga v. Pena-Irala, 630 F.2d 876, 890 (2d Cir. 1980). 55. See, e.g., Landsman, supra note 50, at 90 (arguing that genocide must be prosecuted and never amnestied, because the “interests of a single nation should never be allowed to thwart prosecution”); Hall, supra note 42, at 28; Ian Martin, Justice and Reconciliation: Responsibilities and Dilemmas of Peace-makers and Peace-builders,” in THE LEGACY OF ABUSE: CONFRONTING THE PAST, FACING THE FUTURE 3, 82 (Alice H. Henkin ed., 2001) (arguing that the amnesty is not acceptable for violators of humanitarian law).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

820

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

1.

The Institutional Responsibility Claim

The institutional responsibility claim need not invoke any comprehensive philosophy of punishment, but relies instead on a moral division of labor among governmental institutions. Some of its proponents may accept consequentialist or other justifications for impunity yet believe that it is no business of the ICC to consider them. Rather, political institutions should do politics and policy; judicial institutions should do justice.56 So the Rome Statute leaves political judgments to the Security Council, which is authorized to defer ICC prosecutions when they are ill-advised;57 the ICC itself must not consider political exigencies lest it betray its specific mission—to end impunity. Or, slicing the division of labor even thinner, even if the ICC judiciary may take political factors into account, this is not the role of the prosecutor. I want to note two links in this argument that require support before it can be taken to underwrite a policy that excludes the political situation and its consequences from prosecutorial consideration. First, this argument depends on an interpretation of
56. John Rawls argued that a similar division of labor might apply to punishment more generally, stating that “the judge and the legislator stand in different positions and look in different directions: one to the past, the other to the future.” John Rawls, Two Concepts of Rules, 64 PHIL. REV. 3, 6 (1955). Rawls continued, “The justification of what the judge does, qua judge, sounds like the retributive view; the justification of what the (ideal) legislator does, qua legislator, sounds like the utilitarian view.” Id. With regard to the ICC, see the comments of Michael Reisman, arguing that the ICC is a judicial body that must leave politics to the Security Council, and Beatrice Le Fraper Du Hellen, arguing the opposite. Round Table: Prospects for the Functioning of the International Criminal Court (May 15, 1999), in THE ROME STATUTE OF THE INTERNATIONAL CRIMINAL COURT: A CHALLENGE TO IMPUNITY 291, 299–301, 304–07 (Mauro Politi & Guiseppe Nesi eds., 2001) [hereinafter Round Table]. 57. Article 16 of the Rome Statute authorizes the Security Council to defer an ICC prosecution for renewable periods of a year. See supra note 1, art. 16. However, it can only do so by issuing a resolution under Chapter VII of the U.N. Charter, which has its own requirements. Chapter VII mechanisms are only available in response to a threat to the peace, a breach of the peace, or an act of aggression. U.N. Charter art. 39. This requirement, however, has been interpreted quite expansively to include interventions on behalf of human rights. See, e.g., Roht-Arriaza, supra note 48, at 80. Some question whether even the Security Council should entertain exchanging justice for peace, as it might through repeated deferrals of ICC action. Id. (noting that the Charter requires Chapter VII actions to be consistent with respect for human rights and questioning whether some deferrals would run afoul of this provision). Ian Martin notes that in cases of conflict between peace and justice, the U.N. “has a special responsibility as the guardian of the Charter and of international law. . . . [I]t cannot condone amnesties regarding war crimes, crimes against humanity and genocide, or foster those which violate relevant treaty obligations of the parties.” Martin, supra note 55, at 82. Of course, a Security Council deferral would not come within this proscription, at least so long as it is not renewed to an extent that would be tantamount to amnesty.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

821

the ICC’s mission, which is itself quite contestable. ICC Deputy Prosecutor Serge Brammetz says that one crucial and still unresolved question is whether the ICC is a political system with judicial tools or a judicial system surrounded by a political reality.58 For example, when ICC investigation and peace efforts are proceeding simultaneously, should the prosecutor communicate with peace negotiators, government officials, NGOs, donor states, and others as to how he can utilize his powers to assist rather than hinder peace prospects? Brammetz himself believes the ICC is “here for justice, not politics. . . . The priority of the Rome Statute is to prosecute[;] it’s not to provide political stability.”59 And there are indeed reasons why this institutional role makes sense. Most crucially, the ICC lacks the expertise to engage in post-conflict management or peacemaking, or even to accurately assess the sincerity of peace proposals. But it does have the unique potential, underscored by its complementarity principle, to act as a back-up mechanism removed from the political pressures that typically prevent states from prosecuting.60 If the ICC also finds a prosecution too costly to pursue based on these same pressures, it would merely be replicating the problem the ICC was created to solve. Nevertheless, the Rome Statute’s Preamble conceives the ICC as an instrument designed both to “put an end to impunity” and to contribute to the prevention of “grave crimes [that] threaten the peace, security and well-being of the world.”61 The Chairman of the Rome Diplomatic Conference, Phillippe Kirsch, says that the statute was designed to leave open the possibility of ICC deference to a state’s amnesty,62 and others who were there claim that it was specifically intended to factor political exigencies into its decisions.63
58. Interview with Serge Brammetz, Deputy Prosecutor, ICC, at the Hague, Neth. (May 27, 2004). 59. Id. 60. See THOMAS W. POGGE, WORLD POVERTY AND HUMAN RIGHTS: COSMOPOLITAN RESPONSIBILITIES AND REFORMS 150 (2002) (noting that “actors who optimize locally, at each decision point, may not succeed in optimizing globally, over time”). 61. Rome Statute, supra note 1, pmbl. The tension between the retributive and preventive goals of the ICC is likely to be resolved as the ICC proceeds to construe Article 54, the interests-of-justice provision, and Article 17(1)(d), which provides that a case is inadmissible if of insufficient “gravity.” 62. The statement was made to Michael Scharf in 1998. Scharf, supra note 3, at 521– 22, nn.104–06. Others take a similar position. See Gerhard Hafner et al., A Response to the American View as Presented by Ruth Wedgwood, 10 EUR. J. INT’L L. 108, 109–10 (1999); Roht-Arriaza, supra note 48, at 79–80. But see Wedgwood, supra note 3, at 96–97. 63. See, e.g., Round Table, supra note 56, at 299, 301, 305 (comments of Du Hellen), 306 (comments of Kirsch arguing that the court will not commit political suicide by ignoring political reality); Hall, supra note 42, at 7–8; Brubacher, supra note 40, at 93–94 (arguing that the prosecutor must take into account political factors of international peace and

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

822

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

Moreover, sometimes pursuing one case now will damage the prospects of bringing many others later. For the Court to end impunity over the long term, it must assure its own efficacy, and take care to safeguard that of responsible states; as Judith Shklar argued, sometimes an overly legalistic program can destroy “the very social circumstances which condition its effect and even its possibility.”64 Had there been no South African amnesties, there might be no legitimate legal system in place today, given their centrality in obtaining a peaceful transition from the apartheid system. The second, further link to be established is this: Even if one assumes that the ICC’s mandate is to do justice and ignore its consequences, why should this mandate be deemed to generate an absolute and impermeable division of labor that insulates the prosecutor from any concern with the external consequences of his decisions? Legal and moral considerations counsel otherwise. The legal point is that the Rome Statute provides a mechanism for the prosecutor to decline a case in the “interests of justice,”65 and, recognizing that victims will be affected by charging decisions, affords them an opportunity to argue against the prosecutor’s proprio motu decision to pursue a case.66 The moral point arises when the other institutions fail to fulfill their roles. If big-power vetoes or other dynamics prevent the Security Council from responding to potentially tragic exigencies, the ICC’s complementary institutional role is itself thrown into question (as was NATO’s obligation under
security). 64. JUDITH N. SHKLAR, LEGALISM 146 (1964). Jack Snyder and Leslie Vinjamuri argue that: Preventing atrocities and enhancing respect for the law will frequently depend on striking politically expedient bargains that create effective political coalitions to contain the power of potential perpetrators of abuses . . . . Amnesty should therefore be recognized as a legitimate tool when it serves the broader interest in establishing the rule of law. Snyder & Vinjamuri, supra note 2, at 5, 6, 14; see also Catherine Lu, The International Criminal Court as an Institution of Moral Generation: Problems and Prospects 10 (Dec. 7, 2004) (unpublished manuscript, on file with author) (arguing that the Rome Statute “leaves moral space open for setting aside the prosecutorial task when its pursuit would likely undermine the conditions or prospects for preventing future violations”). 65. Rome Statute, supra note 1, art. 53(2)(c). However, a decision not to investigate a case on this basis is treated as an exception to the “duty to investigate.” Id. art. 53(1). 66. Under the Rome Statute, victims may submit observations to the Court with respect to admissibility of a case, may communicate views and concerns arising out of their personal interests, and may apply to have a legal representative participate in the proceedings. Id. arts. 19(3), 68(3); see also INT’L CRIM. CT. R. PROC. & EVID. 89 (providing for participation of victims); id. at 91 (providing for the participation of legal representatives of victims), available at http://www.un.org/law/icc/asp/1stsession/report/english/part_ii_a_e.pdf. The victims’ interests are also a factor in an interests-of-justice determination. Rome Statute, supra note 1, arts. 53(1)(c), (2)(c).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

823

the U.N. Charter to cede military action to the United Nations when the Security Council arguably evaded its responsibilities in Kosovo).67 At that point there really is no division of labor at all, and to insist the prosecutor pretend that there is, that he not consider the damage a prosecution might cause even when no one else is either, places him in a morally untenable position. There is a long literature on the proper course when following role-based ethical rules would lead to immoral results, including the famous Milgram experiment, and much legal scholarship on the professional responsibility of both lawyers and judges.68 Commentators on the ICC have only begun to explore how far the ICC’s special institutional role should override more general moral considerations.69 I do no more here than note that the moral division of labor argument rests on some assumptions that we may not be ready to accept. 2. The Consequentialist Claim

The other two major arguments for a strict ICC policy of prosecution and punishment are broader; indeed, they reflect conflicting theoretical claims about what moral criteria should govern decision-making in general. The consequentialist claim says that one should do whatever will produce the best outcome, impartially

67. See Sean Murphy, The Intervention in Kosovo: A Law Shaping Incident?, 94 AM. SOC’Y INT’L L. 302 (2004). 68. STANLEY MILGRAM, OBEDIENCE TO AUTHORITY: AN EXPERIMENTAL VIEW 73–88 (1974) (presenting selected transcripts of an experiment in which most subjects were willing to inflict excruciating pain on another if ordered to do so by a scientist with apparent authority). On the problem of disjunctions between general moral imperatives and rolebased requirements, see R.S. Downie, Roles and Moral Agency, 29 ANALYSIS 39 (1968); Gerald Cohen, Beliefs and Roles, 67 PROC. OF THE ARISTOTELIAN SOC’Y FOR THE SYSTEMATIC STUDY OF PHIL. 17 (1966); see generally ROBERT M. COVER, JUSTICE ACCUSED: ANTISLAVERY AND THE JUDICIAL PROCESS (1975) (examining the perceived obligation of anti-slavery judges to uphold slavery laws). 69. See, e.g., Todd Howland & William Calathes, The U.N.’s International Criminal Tribunal: Is It Justice or Jingoism for Rwanda? A Call for Transformation, 39 VA. J. INT’L L. 135, 146–47 (1998) (arguing that International Criminal Tribunal for the Former Yugoslavia (ICTY) should substitute a human rights orientation for its narrow prosecutorial vision); Round Table, supra note 55, at 299–301 (comments of Du Hellen); Garth Meintjes, Domestic Amnesties and International Accountability, in INTERNATIONAL CRIMES, PEACE, AND HUMAN RIGHTS: THE ROLE OF THE INTERNATIONAL CRIMINAL COURT 83, 89 (Dinah Shelton ed., 2000) (arguing that international tribunals should avoid the politicization that would result from taking consequences of prosecutions into account); Snyder & Vinjamuri, supra note 2, at 44 (“[D]ecisions to prosecute should be taken by political authorities, such as the UN Security Council or the governments of affected states, not by judges who remain politically unaccountable.”).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

824

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

considered.70 The deontological claim insists instead that some acts are themselves right or wrong, independent of whatever outcomes they produce; the ends do not always justify the means.71 In the fields of crime and punishment, some deontologists perceive a strict retributive imperative to punish the guilty regardless of the impact of doing so.72 The contrary consequentialist argument puts all the justificatory weight on the results the punishment is expected to produce. The consequentialist argument is that an appropriately structured punishment will best produce a net social benefit, in excess of the opportunity costs, inmate’s suffering, and other losses it will impose. This is an empirical claim about the utility of punishment that may hold true generally but be false at a particular time or place. If the total consequences of a particular prosecution would ultimately be dire rather than beneficial, the consequentialist argument for prosecution would seem to fail.73 Yet consequentialists are among the strongest advocates for a policy of “prosecution regardless.” They must believe either that predicted dire consequences are so often exaggerated that maximum utility can be obtained by uniformly dismissing such predictions;74 or
70. I use the broader concept, consequentialism, rather than utilitarianism, primarily because the latter is historically associated with particular theories of the good. Consequentialism holds that an action should be judged only by the goodness of its outcome, but does not specify any particular measure of the good. Classical utilitarians generally identified the good with happiness; present-day utilitarians are more likely to identify goodness with aggregate preference satisfaction. See, e.g., John Rawls, Classical Utilitarianism, in CONSEQUENTIALISM AND ITS CRITICS 14, 16–17 (Samuel Scheffler ed., 1988); JEREMY BENTHAM, AN INTRODUCTION TO THE PRINCIPLES OF MORALS AND LEGISLATION (J. H. Burns & H. L. A. Hart eds., Oxford Univ. Press, 1996); RICHARD BRANDT, A THEORY OF THE GOOD AND THE RIGHT (1979). There are several other aspects that distinguish utilitarianism as a subset of consequentialism. See Walter SinnottArmstrong, Consequentialism, in STAN. ENCYC. PHIL. (Edward N. Zalta ed., 2003), available at http://plato.stanford.edu/archives/sum2003/entries/consequentialism. 71. See, e.g., ROBERT NOZICK, PHILOSOPHICAL EXPLANATIONS 494 (1981); CHARLES FRIED, RIGHT AND WRONG 9 (1978) 72. See infra notes 92, 107–13. 73. Precisely that logic was advanced by President Julio Sanguinetti of Uruguay in justifying an amnesty (later ratified by a narrow majority in a popular vote), who argued: The severity of the law in prescribing a penalty must be attenuated when its application would result in a greater social ill than that of allowing a crime to go unpunished. . . . In a word, the renunciation of the power of punishment is merely another way of administering justice, since . . . the intention underlying both of these elements, amnesty and punitive authority, is ultimately to bring peace and tranquility to all members of the community. NEIER, supra note 44, at 97. 74. Leila Nadya Sadat comes close to this position in arguing that even if there are highly exceptional cases in which amnesty would avoid disaster in the specific case, “the exception easily becomes the rule, with highly corrosive effects on the rule of law, as well as

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

825

more likely, that the immediate consequences, although dire, will be ultimately outweighed by long-term benefits. However, there is a major difficulty, arguably not insurmountable, with identifying and calculating the costs and benefits. In most domestic systems, the expected benefit will usually involve a reduction in crime, due to the deterrent, norm-reinforcing, incapacitative, or rehabilitative impact of incarcerating offenders. For the ICC, this is also an expectation. Many people argue that if genocide and crimes against humanity are reliably prosecuted and punished, a deterrent effect should follow—if not for some megalomaniacal leaders, at least for the subordinates who would otherwise carry out their orders.75 However, for the ICC there may be additional costs and benefits that are more wide-ranging and unpredictable. In the LRA case, for example, ICC charges might help end the insurrection by stigmatizing and marginalizing Kony and other leaders, or they might prolong it by deterring them from surrender;76 prosecutions might lead rebels either to flee the LRA, or to kill prospective witnesses.77
international peace and security.” Sadat, supra note 48, at 34. Although not without ambiguity, she seems to argue that such a case appears so rarely that even when confronted by one, more damage would be done by accepting amnesty in that case than by accepting the consequences of prosecution. Id. A contrary view is offered by political scientists Jack Snyder and Leslie Vinjamuri. Based on their study of thirty-two countries coping with the aftermath of civil wars between 1989 and 2003, Snyder and Vinjamuri concluded that “amnesty can be an indispensable tool in reaching peace settlements when perpetrators remain strong. Moreover, amnesty per se does not appear to be incompatible with the subsequent consolidation of peace.” Snyder & Vinjamuri, supra note 2, at 39. Similar arguments have recently proliferated on the subject of torture and other coercive interrogation techniques, with some arguing that even if sometimes justified, the cases are so rare that a flat prohibition is optimific, and others arguing that there is no empirical evidence to support that view. See, e.g., Henry Shue, Torture, 7 PHIL. AND PUB. AFF. 124 (1978) (advocating a ban); Owen Gross, Are Torture Warrants Warranted? Pragmatic Absolutism and Official Disobedience, 88 MINN. L. REV. 1481, 1486–87 (2004) (advocating a ban); Adrien Vermeule & Eric Posner, Should Coercive Interrogation Be Legal?, 104 MICH. L. REV. 671 (2006) (arguing for case-by-case determinations). 75. See, e.g., Payam Akhavan, Beyond Impunity: Can International Criminal Justice Prevent Future Atrocities?, 95 AM. J. INT’L L. 7, 9–13 (2001); Parliamentarians for Global Action, A Deterrent International Criminal Court—The Ultimate Objective (Dec. 6–7, 2004), available at http://www.iccnow.org/conferencesmeetings/reportsdeclarations/ asiaoceaniareports/2004/WellingtonPGAPaperonICCDeterrence7Dec04.pdf. 76. See, e.g., Akhavan, supra note 75, at 12 (stating trials can stigmatize and undermine genocidal politicians). The ICC focus on Uganda has already encouraged Sudan’s withdrawal of support from the LRA. Akhavan, supra note 17, at 404. More generally, to the degree that African state budgets are significantly dependent on foreign donors, pressure from donor states can be effective, and ICC charges are likely to trigger such pressure. 77. For example, over three hundred Rwandan survivors scheduled to testify in

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

826

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

In countries recovering from ethnic wars, ICC prosecution might help reconcile hostile ethnic groups by tying atrocities to individuals and bringing them to justice, or instead fuel a further cycle of vengeance if deemed an instrument of one side.78 A fragile post-conflict democracy that cooperates with ICC prosecution of its predecessors might strengthen its legitimacy and the rule of law,79 or appear to be a powerless tool of the West,80 or even motivate a coup d’etat. With
genocide trials were murdered between 1994 and 1997. Jennifer Widner, Courts and Democracy in Postconflict Transitions: A Social Scientist’s Perspective on the African Case, 95 AM. J. INT’L L. 64, 67–68 (2001). 78. In urging the United Nations to establish an international criminal tribunal for Rwanda, Rwanda’s U.N. ambassador argued that “the trial of those accused of serious breaches of international humanitarian law and acts of genocide by an external impartial body would help promote peace and reconciliation among the parties and contribute to the stabilization of the situation in Rwanda.” The Secretary-General, Progress Report of the Secretary-General on the United Nations Assistance Mission for Rwanda, U.N. Doc. S/1994/1133 ¶ 11 (Oct. 6, 1994); see also Diane Orentlicher, Swapping Amnesty for Peace and the Duty to Prosecute Human Rights Crimes, 3 ILSA J. INT’L & COMP. L. 713 (1997); Neil Kritz, Where We Are and How We Got Here: An Overview of Developments in the Search for Justice and Reconciliation, in THE LEGACY OF ABUSE: CONFRONTING THE PAST, FACING THE FUTURE 21, 27–29 (Alice H. Henkin ed., 2002) (comparing national to international prosecutions, and attributing ethnic wars in the former Yugoslavia to previous attempts to suppress knowledge and justice); Furman v. Georgia, 408 U.S. 238, 308 (1972) (Stewart, J., concurring) (stating that when people “begin to believe that organized society is unwilling or unable to impose upon criminal offenders the punishment they ‘deserve,’ then there are sown the seeds of anarchy—of self-help, vigilante justice, and lynch law”). Whether trials and punishments can have this effect depends crucially on the credibility and impartiality of the system imposing justice, however. Early in its operations, the ICTR was criticized for pursuing Hutu but not Tutsi criminals, which some commentators believe undermined Rwandan reconciliation. See Helena Cobban, Healing Rwanda: Can an International Court Deliver Justice?, 28 BOSTON REV., Dec. 2003/Jan. 2004, available at http://bostonreview.net/BR28.6/cobban.html. The same charge has been made against the ICTY. See, e.g., Snyder & Vinjamuri, supra note 2, at 21 (arguing that the ICTY has reinforced ethnic cleavages, with Serbs and Croats each complaining that their group has been singled out for prosecutions). 79. For claims that this dynamic prevails generally, see Sadat, supra note 48, at 9 (arguing that “amnesty deals typically foster a culture of impunity in which violence becomes the norm, rather than the exception”); Neil J. Kritz, War Crime Trials: Who Should Conduct Them—and How?, in WAR CRIMES: THE LEGACY OF NUREMBERG 168, 175 (Belinda Cooper ed., 1999); Landsman, supra note 50, at 83–84; Diane Orentlicher, Settling Accounts: The Duty to Prosecute Human Rights Violations of a Prior Regime, 100 YALE L.J. 2537, 2542–44 (1991) (summarizing the case for prosecutions, including their ability to promote rule of law, democratic legitimacy, and civilian dominance); RUTH TEITEL, TRANSITIONAL JUSTICE 28–30 (2000); CARLOS SANTIAGO NINO, RADICAL EVIL ON TRIAL (1996) (“[R]etroactive justice for massive human rights violations helps protect democratic values. An aggressive use of the criminal laws will counteract a tendancy towards unlawfulness, negate the impression that some groups are above the law, and consolidate the rule of law.”). But see Snyder & Vinjamuri, supra note 2, at 43 (finding that trials “are not highly correlated with the consolidation of peaceful democracy”). 80. See, e.g., Snyder & Vinjamuri, supra note 2, at 22 (arguing that the ICTY did not

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

827

regard to any particular case, everything depends on the specific situation and the projected consequences of alternative scenarios. The daunting problem with consequentialist arguments is that such accurate cost-benefit predictions are as elusive as they are crucial. One can only attempt to induce an answer from the historical record, and this process is inherently contestable. It is notoriously difficult to determine whether a past regime of punishments had subsequent deterrent effects.81 It is all the more difficult to predict the future, where the future that counts involves agents and circumstances far removed from past precedent. Moreover, the calculation of costs and benefits must extend well beyond the immediate consequences to the long term effects of a prosecution (or a mandatory prosecution rule).82 For example, it could well be the case that bringing charges against Kony and other LRA leaders will result in prolonging the casualties and atrocities, but that doing so will establish a credible deterrent that would curtail many more such tragedies in the future. But if, as social scientists say, we lack the tools to predict whether a thoroughly examined individual will commit a future violent act,83 how can we know whether a series of
help strengthen legal institutions in the post-Yugoslav successor states, and citing a survey finding that Bosnian judges and prosecutors felt marginalized by the ICTY because of its location in the Hague, the lack of communication between Bosnian and tribunal legal professionals, criticisms of the Bosnian legal system made by the international legal community, and the perceived political nature of the tribunal); Kristin Cibelli & Tamy Guberek, Justice Unknown, Justice Unsatisfied? Bosnian NGOs Speak about the International Criminal Tribunal for the Former Yugoslavia 14–16 (2000), available at http://www.epiic.com/class/justicereport.pdf. 81. Certainly research on deterrence proceeds according to rigorous scientific standards, but there are few if any deterrent studies that have inspired universal agreement. Of course, unlike laboratory research with its random assignments into treatment and control groups, deterrence researchers have no access to an otherwise identical specimen, here, a society, to serve as a control group. Despite the use of regression analysis and other statistical tools, observational research does not itself prove a causal hypothesis, because the correlations it reveals can always be the result of some unknown, unseen cause. For an account of the many criticisms lodged against earlier and contemporary death penalty deterrent studies, see Robert Weisberg, The Death Penalty Meets Social Science: Deterrence and Jury Behavior Under New Scrutiny, 1 ANN. REV. L. & SOC. SCI. 151 (2005), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=752044; John J. Donahue and Justin Wolfers, Uses and Abuses of Empirical Evidence in the Death Penalty Debate, 58 STAN. L. REV. 791 (2005). 82. See J.C.C. Smart, An Outline of Utilitarian Ethics, in UTILITARIANISM: FOR AND AGAINST 1, 40–41 (J.C.C. Smart & Bernard Williams eds.,1973) (stating that utilitarianism needs and lacks some method of assigning numerical probabilities to any imagined future event). 83. See John Monahan, The Prediction of Violent Behavior: Toward a Second Generation of Theory and Policy, 141 AM. J. PSYCHIATRY 10, 11 (1984) (fewer than half of mental health professionals’ predictions of dangerousness proved accurate); Barefoot v.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

828

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

ICC prosecutions will deter a future would-be tyrant, or just make him cling to power by any means necessary?84 Without sufficiently compelling social science, a prosecutor who wants to evaluate the ultimate consequences of a decision to prosecute lacks critical information with which to do so. Perhaps the opposing reactions people have to ICC involvement in Uganda more often reflect different visions of its outcome than different moral premises. The most optimistic vision sees a consistent commitment to prosecuting war criminals as the one route to a new international culture, one in which international politicians take the rule of law more seriously and even dictators find some kinds of brutality and atrocity unthinkable.85 Others believe
Estelle, 463 U.S. 880, 901 n.7 (1983) (citing early work by John Monahan finding that twothirds of individuals predicted to be violent were false positives); Paul J. Robinson & John M. Darley, The Utility of Desert, 91 NW. U. L. REV. 453, 465 (1997) (citing study showing forty-seven percent of those predicted to recidivate did not do so). But see Christopher Slobogin, The Civilization of the Criminal Law, 58 VAND. L. REV. 121, 144–45 (2005) (arguing that predictions are now sufficiently accurate). 84. There are several problems that arise in applying the deterrent theory to crimes against humanity, including the lack of empirical data on the effect of international trials and the poor applicability of theories of rational choice in the context of either collective violence or extremely powerful individuals. See, e.g., Charles Villa-Vicencio, Why Perpetrators Should Not Always Be Prosecuted: Where the International Court and Truth Commissions Meet, 49 EMORY L.J. 205, 210 (2000) (stating that “fixed mind-sets, entrenched prejudice and the kind of ideological bloody-mindedness that drives militant perpetrators of political crime are forces too powerful to be curtailed by the threat of future prosecution alone”). Moreover, insofar as prosecutions are uncertain, unsuccessful, or highly selective—conditions historically likely in the case of crimes against humanity—any deterrent effect they might have would be undercut. See, e.g., Mark Drumbl, Punishment, Postgenocide: From Guilt to Shame to Civis in Rwanda, 75 N.Y.U. L. REV. 1221, 1255 (2000) (“[T]he uncertainty of sanction is more determinative than the severity of sanction in influencing behavior.”); MARTHA MINOW, BETWEEN VENGEANCE AND FORGIVENESS 50 (1998) (quoting the Nuremberg prosecutor Robert Jackson as stating that personal punishment “is probably not to be sufficient deterrent to prevent a war where the war-makers feel the chances of defeat to be negligible”). The Snyder-Vinjamuri study found that neither the ICTY nor the ICTR have deterred atrocities, citing the sustained pace of atrocities after indictments and trials, both in the regions and globally. Snyder & Vinjamuri, supra note 2, at 20, 24–25. But see Zeidy, supra note 5, at 944 n.301 (quoting Kofi Annan as stating that “ending of impunity is a vital prerequisite for post-conflict peace-building”). Regarding the deterrent effect of the ICC in particular, see Parliamentarians for Global Action, supra note 75; David A. Wippman, Exaggerating the ICC 169–82 (Cornell L.. Sch. Res. Paper No. 04018), available at http://ssrn.com/abstract=601747; Anne-Marie Slaughter, Not the Court of First Resort, WASH. POST, Dec. 23, 2003, at B7; Robinson, supra note 35, at 482. 85. See, e.g., NEIER, supra note 44, at 103–04; Kritz, supra note 50, at 128 (predicting that prosecutions will establish a new societal dynamic of accountability). For more general studies on the power of law to convey and reinforce certain norms, see, for example, Dan M. Kahan, Social Influence, Social Meaning, and Deterrence, 83 VA. L. REV. 349 (1997); Lawrence Lessig, Social Meaning and Social Norms, 144 U. PA. L. REV. 2181 (1996); Richard H. McAdams, The Origin, Development, and Regulation of Norms, 96 MICH. L.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

829

this is a utopian vision that must be severely discounted by its uncertainty, and cannot justify decisions that in the meantime would prolong a war or a tyrannical regime.86 A consequentialist prosecutor’s unavoidable but extraordinarily difficult task is to make decisions that invoke such magnificent hopes and terrible costs with so little predictive information. 3. The Retributive Justice Claim

A standard criticism of the consequentialist criterion is that it attaches no importance to justice and desert, except on the derivative basis that either might affect the utility of punishment.87 Because consequentialism posits aggregate utility as the ultimate criterion, a sufficiently beneficial outcome would justify imposing draconian punishments for minor offenses, and a sufficiently detrimental outcome would require dropping the case.88 The retributive argument for prosecution and punishment would seem less vulnerable to pleas to trade justice for peace because its criterion is not aggregate social benefit but justice itself. Broadly construed, retributivism holds that one who freely chose to victimize another deserves to suffer accordingly, and/or one who was victimized deserves rectification. When the victimization is criminal, punishment usually has been taken to be the necessary means to accomplish this, its imposition and degree determined by looking backwards to their just deserts.89 The future consequences of
REV. 338 (1997). 86. See, e.g., Snyder & Vinjamuri, supra note 2, at 5 (“[T]he prosecution of perpetrators of atrocities according to universal standards . . . risks causing more atrocities than it would prevent, because it pays insufficient attention to political realities.”). 87. For example, Paul H. Robinson and John M. Darley argue that departing from the desert-based criteria that people expect would so undermine the credibility of the criminal law, and thus its norm-reinforcing effects, as to outweigh the benefits of doing so. Robinson & Darley, supra note 85, at 456–67; see also TOM R. TYLER, WHY PEOPLE OBEY THE LAW 168–69 (1990) (finding that law abiding behavior does not respond “primarily to reward and punishment [but to] the legitimacy of legal authorities and the morality of the law”); cf. H.L.A. HART, Prolegomenon to the Principles of Punishment, in PUNISHMENT AND RESPONSIBILITY: ESSAYS IS IN THE PHILOSOPHY OF LAW 12 (1968). 88. See, e.g., Samuel Scheffler, Introduction to CONSEQUENTIALISM AND ITS CRITICS 1, 2 (Samuel Scheffler ed., 1988) (citing an argument in JOHN RAWLS, A THEORY OF JUSTICE (1971) that utilitarianism is indifferent as to how satisfactions and dissatisfactions are distributed among distinct individuals); DAVID A.J. RICHARDS, THE MORAL CRITICISM OF LAW 232–33 (1977). 89. Desert is usually taken to be a function of the gravity of the act and the blameworthiness of the actor, the latter requiring an inquiry into the actor’s intent, situation, and capacity. As C. S. Lewis wrote: [T]he concept of Desert is the only connecting link between punishment and

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

830

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

punishment are irrelevant in two ways: The reason for inflicting it has nothing to do with any expected social benefit, and (in the mainstream version) deserved punishment must be dispensed regardless of its social cost. The retributive idea as applied to crimes has been defined in various ways. The version that fuels the demand for prosecution without regard to subsequent outcomes is what I will label strict retributivism. Let us stipulate that strict retributivism includes three principles:90 (1) A restrictive principle, prohibiting punishment of the innocent and of the guilty beyond what they deserve. There might be great deterrent value in inflicting punishment on a mobster’s innocent mother, but doing so would be intrinsically unjust, a violation of the Kantian moral imperative to treat a person always as an end in himself, and never as a means only.91
justice . . . . [W]hen we cease to consider what the criminal deserves and consider only what will cure him or deter others, we have tacitly removed him from the sphere of justice altogether; instead of a person, a subject of rights, we now have a mere object, a patient, a “case.” C. S. Lewis, The Humanitarian Theory of Punishment, in PHILOSOPHY AND CONTEMPORARY ISSUES 71, 72 (John R. Burr & Milton Goldinger eds., 1972). The concept of desert most centrally embodies respect for the right of autonomous individuals to determine their futures. On desert as a basis for punishment, see generally Edward Rubin, Just Say No to Retribution, 7 BUFF. CRIM. L. REV. 17, 27–36 (2003); JOEL FEINBERG, DOING AND DESERVING: ESSAYS IN THE THEORY OF RESPONSIBILITY (1970); Sanford H. Kadish, Why Substantive Criminal Law—A Dialogue, 29 CLEV. ST. L. REV. 1, 10–11 (1980); JOHN KLEINIG, PUNISHMENT AND DESERT (1973); Herbert Morris, Persons and Punishment, 52 MONIST 475 (1968); John Cottingham, Varieties of Retribution, 29 PHIL. Q. 238 (1979). 90. Some less orthodox retributivists would deny that the second or third principles are necessary to the retributive justice theory. So-called limiting retributivists reject the second principle, believing that crime is a necessary but not sufficient reason for punishment. See infra note 105. Others reject the third principle, holding instead that retributive justice need not involve the state, which is certainly true—just and unjust punishments can exist within families, for example—but controversial when applied to crimes. See, e.g., Jean Hampton, Correcting Harms Versus Righting Wrongs: The Goal of Retribution, 39 UCLA L. REV. 1659, 1694 (1992) (“[T]here are many different persons or institutions that are possible agents of retribution.”); cf. James Q. Whitman, What is Wrong with Inflicting Shame Sanctions?, 107 YALE L.J. 1055, 1059, 1088–89 (1998) (arguing that as a matter of liberal political theory, if not punishment theory, shame sanctions are impermissible because they delegate the state’s responsibility to deliver measured punishment to people whose reactions are not controllable by the state). For a description of nine distinct claims that are sometimes put forward as retributivism, see generally Cottingham, supra note 89. I concentrate on the strict retributivist version because it is the only one that comports with the argument that the ICC must pursue prosecutions without regard to the consequences of doing so. 91. See IMMANUEL KANT, THE METAPHYSICS OF MORALS 105 (Mary Gregor trans., Cambridge University Press 1996), in which he applies one formulation of the categorical imperative (“the formula of humanity”) to punishment as follows: “Punishment by at court

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

831

(2) An affirmative duty, that punishment must be inflicted on the guilty, and in proportion to their desert;92 and (3) A procedural principle intended to insulate retribution from vengeance, that the punishment must be inflicted according to law by the state rather than by the victim of the crime.93 Psychological satisfaction to the victim is merely incidental. The important principle for our purposes is the second one, that crime demands punishment. The principle taken in earnest is quite exacting. It requires that punishment neither exceed what is deserved nor fall short, and this implies that considering any factor other than desert invites injustice, in the form of undeserved prosecutions, deserved prosecutions forgone, or disproportionate sentences.94 Is there such an affirmative duty, and how would we know this? For many people, the duty to bring the worst criminals to justice is a deep sentiment, or an article of faith. Some theorists believe that is enough, that there need not be any further justification.95 This becomes an increasingly difficult position to
. . . can never be inflicted merely as a means to promote some other good for the criminal himself or for civil society.” Id. For opposing views, see J. C. C. Smart, An Outline of a System of Utilitarian Ethics, in UTILITARIANISM: FOR AND AGAINST 1, 69–72 (J. C. C. Smart & Bernard Williams eds., 1973); Mirko Bagaric & Kumar Amarasekara, The Errors of Retributivism, 24 MELB. U. L. REV. 124 (2000); Cass R. Sunstein & Adrian Vermeule, Is Capital Punishment Morally Required? Acts, Omissions, and Life-Life Tradeoffs, 58 STAN. L. REV. 703 (2005). 92. Most contemporary retributivists agree that the theory requires punishment (although not all, as described infra at note 105). For a justification of this version of retributivism, see Michael S. Moore, The Moral Worth of Retribution, in RESPONSIBILITY, CHARACTER, AND THE EMOTIONS: NEW ESSAYS IN MORAL PSYCHOLOGY 179 (Ferdinand Schoeman ed., 1987). Moore argues that “the distinctive aspect of retributivism is that the moral desert of an offender is a sufficient reason to punish him or her.” Id. at 180. 93. See, e.g., ERNEST VAN DEN HAAG, PUNISHING CRIMINALS: CONCERNING A VERY OLD AND PAINFUL QUESTION 10–11 (1975); ROBERT NOZICK, PHILOSOPHICAL EXPLANATIONS 367 (1981) (both arguing that retribution is distinct from vengeance, which is personal). The expectation that infliction by the state will insulate retribution from vengeance may be overly optimistic. See James Q. Whitman, Making Happy Punishers, 118 HARV. L. REV. 2698 (2005) (book review). 94. This stringent requirement of precision led Kant to the conclusion that only the principle of lex talionis, colloquially translated as “an eye for an eye,” would conform to retributive principles. “[O]nly the law of retribution (jus talionis) . . . can specify definitely the quality and the quantity of punishment; all other principles are fluctuating and unsuited for a sentence of pure and strict justice . . . .” KANT, supra note 91, at 105–06. 95. See, e.g., JAMES FITZJAMES STEPHEN, A GENERAL VIEW OF THE CRIMINAL LAWS OF ENGLAND 99 (1863) (stating that a criminal conviction confirms and justifies the sentiment of hating criminals: “The criminal law stands to the passion of revenge in much the same relation as marriage to the sexual appetite”); Jeffrie G. Murphy, The Retributive Emotions, in FORGIVENESS AND MERCY 1, 6 (Jeffrie G. Murphy & Jean Hampton eds., 1988), Jeffrie G.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

832

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

maintain, however, as the costs of proceeding with a prosecution become larger. We need to scrutinize the principle more deeply, in two respects that it is helpful to keep separate: Is there a (nonconsequentialist) moral duty to prosecute and punish egregious crime? And, assuming there is such a duty, is it absolute rather than negotiable in the face of other considerations? The retributivist who would have the prosecutor bring charges regardless of their effects on third parties requires an affirmative answer to both questions. These questions are the subjects of the next section. B. When Peace and Prosecution Conflict

Here I consider how the claims of retributive justice should apply in a case where an ICC prosecution would have a dire impact on third parties. The Acholis’ warning that ICC charges against LRA leaders would prolong the war is just the latest in a long line of such claims. Of course, it is easy to exaggerate the threat, and in transitional situations there have been many claims that prosecution would reap disaster, which have proven unfounded.96 The Acholi claim may prove to be one of them. Nevertheless, no one should underestimate the prospect of genuine “peace versus punishment” dilemmas.97 One might have emerged had Saddam been willing to
Murphy, Hatred: A Qualified Defense, in FORGIVENESS AND MERCY 88, 108 (Jeffrie G. Murphy & Jean Hampton eds., 1988) (retributive hatred as an appropriate response to wrongdoing). Is there really no need for reasoned justification? A moral imperative to inflict suffering on the blameworthy is not easily justified simply by showing it is deeply entrenched in our feelings, for example; one must show why we should honor these feelings. Michael Moore tries to take feelings of guilt and contrition after wrongdoing to this step, arguing that these emotions are epistemologically valid heuristics that can lead us to the moral truth of retribution. See Moore, supra note 92, at 179. But see Cottingham, supra note 89, at 239 (noting that justifying punishment by desert “reduces to the bald assertion that it is simply just that the offender should be punished”); Anthony R. Duff, Crime and Punishment, in ROUTLEDGE ENCYCLOPEDIA OF PHILOSOPHY (E. Craig ed., 1998), available at http://www.rep.routledge.com/article/T002 (“[A]ll retributivist theories [fail] to explain this notion of penal desert, falling back on unexplained intuition or metaphysical mysterymongering.”). 96. Orentlicher, supra note 79, at 2548 n.37 (citing examples of exaggerated predictions). 97. Despite great advances in international criminal accountability in the last decade, see supra note 4, amnesties and/or safe passage to asylum in exile have been granted to secure peace or democratic transitions too often to discount the reality of the dilemma in at least some cases. Such was the case in post-Communist Eastern Europe, post-Franco Spain, Cambodia, South Africa, Chile, Brazil, Guatemala, Argentina, Uruguay, Nicaragua, Peru, El Salvador, Haiti, Philippines, Uganda, and many more. See Scharf, supra note 3, at 507; NEIER, supra note 44, at 97–98; NINO, supra note 79, at 17; Orentlicher, supra note 79, at 2544–46. Jack Snyder and Leslie Vinjamuri argue more generally that “in transitional countries . . .

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

833

accept President Bush’s forty-eight hour “get out of town” ultimatum if provided asylum.98 As of this writing, many African leaders are trying to end the long misery of Zimbabwe’s people by coaxing its corrupt leader, Robert Mugabe, into an asylum in exile;99 the British government is seeking to establish a non-incarcerative alternative for handling crimes committed during the “troubles” in Northern Ireland;100 the newly elected president of Liberia is weighing “calls for justice against the need for peace” in determining whether to seek extradition of predecessor Charles Taylor to the tribunal that indicted him;101 and a large majority of Algerians have just voted in favor of granting a blanket amnesty on all but the most heinous crimes, as a step towards reconciliation after over a decade of civil war.102 The ICC is not likely to be fortunate enough to escape daunting dilemmas of this kind. Indeed, two of the three ICC referrals accepted thus far involve potential defendants who are engaged in on-going wars or atrocities, and prosecutorial decisions may fundamentally affect what happens on the ground. For that reason, many ICC supporters have argued that a major aim of the ICC should be cessation—intervening in ways that can help end a

[d]ecisions to try members of the former regime should be weighed against the possibly adverse effects on the strengthening of institutions.” Snyder & Vinjamuri, supra note 2, at 14–15. One such example is Argentina, where prosecutors of the military junta were forced to calculate how far they could go down the ranks without threatening instability or even a military coup. President Alfonsin’s advisors Carlos Nino and Jaime Malamud Goti have each described these calculations. Jaime Malamud-Goti, Transitional Governments in the Breach: Why Punish State Criminals?, 12 HUM. RTS. Q. 1, 5 (1990) (arguing that Argentina’s decision to prosecute only some did not breach its duties of justice); Carlos S. Nino, The Duty to Punish Past Abuses of Human Rights Put into Context: The Case of Argentina, 100 YALE L.J. 2619, 2635, 2638 (1991) (demanding prosecution of every case in Argentina would have hurt the prospects of bringing the leaders to justice). 98. See President George W. Bush, President Says Saddam Hussein Must Leave Iraq within 48 Hours (Mar. 18, 2003), available at http://www.whitehouse.gov/news/releases/ 2003/03/20030317-7.html. The United States is currently urging Iraqis to include former Bathist officials in their government in the interests of peace. Richard A. Oppel, Jr., Rice, in Baghdad, Urges Sunni Role in Constitution, N.Y. TIMES, May 16, 2005, at A1. 99. Robert I. Rotberg, The End of Tyranny in Zimbabwe, BOSTON GLOBE, Aug. 18, 2005, at A15. 100. Kevin Sullivan, In N. Ireland, Some Wounds are Still Raw, BOSTON GLOBE, Dec. 11, 2005, at A43. 101. Lydia Polgreen, Liberia’s Harvard-Trained “Queen” Is Sworn In as Leader, N.Y. TIMES, Jan. 17, 2006, at A3 (stating that after fomenting war and atrocities throughout Liberia and neighboring countries in Africa, warlord-turned-President Taylor was coaxed to depart for Nigeria, which will not deliver him to the hybrid Sierra Leone tribunal unless the Liberian government requests it). 102. Algerians Seen Backing Amnesty for Militants, WASH. POST, Sept. 30, 2005, at A14.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

834

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

conflict.103 But what should happen when ICC non-intervention is important in ending a conflict, as the Acholi leaders allege? For consequentialists, the answer turns on the global, longterm results. It is the retributivist view, that justice must be done for justice’s sake, that some believe implies a duty to prosecute and punish by the state or the ICC despite the results of doing so. In what follows I offer two reasons why the demands of retributive justice may not go so far. First, this obligation to do justice is not identical to, and does not always entail, a simple duty to prosecute and punish. In the aftermath of crime, the essential duty of a state (or upon its default, the ICC) is to recognize and repudiate the crime, and stand in solidarity with the victim. Criminal punishment is ordinarily an effective means of achieving this, but sometimes other instruments may be as well. I believe the South African TRC did so, as argued in Part III. Second (and even if we assume that retributive justice does entail a specific obligation to prosecute and punish), that obligation may not be absolute—not “at any price.” Both arguments suggest that there are some broader and more nuanced options available to the prosecutor who fears extreme consequences would result from a prosecution. 1. Does Justice Always Demand Prosecution?

Is there an affirmative duty to prosecute and punish criminals, and if so, why? Strict retributivists believe that this duty is the essence of retributivism, but another school—call it limiting retributivism—does not include a duty to punish in the theory. It holds that only the blameworthy may be punished, not that they must be; so desert is a necessary but not sufficient condition for punishment.104 Limiting retributivism jettisons the most
103. See, e.g., Gustavo Gallon, The International Criminal Court and the Challenge of Deterrence, in INTERNATIONAL CRIME, PEACE, AND HUMAN RIGHTS: THE ROLE OF THE INTERNATIONAL CRIMINAL COURT 93, 94–97 (Dinah Shelton, ed., 2000). 104. HERBERT L. PACKER, THE LIMITS OF THE CRIMINAL SANCTION 62–66 (1968). Norval Morris is also largely identified with “limiting retributivism,” although he did allow for desert to set a lower limit when necessary to avoid deprecating the seriousness of the crime. See NORVAL MORRIS, THE FUTURE OF IMPRISONMENT 60, 78–79 (1974), discussed in Richard S. Frase, Sentencing Principles in Theory and Practice, 22 CRIME & JUST. 363, 363– 64 (1997); see also Anthony M. Quinton, On Punishment, in THE PHILOSOPHY OF PUNISHMENT 55, 58 (H.B. Acton ed., 1969) (An affirmative duty to punish is “not entailed by the retributivist attack on utilitarianism and has none of the immediate compulsiveness of the doctrine that guilt is the necessary condition of punishment.”); Nino, supra note 97, at

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

835

questionable aspects of both strict retributivism (which mandates suffering even when socially useless105) and consequentialism (which may authorize scapegoating of the innocent if socially useful). What remains is an account that combines the overwhelming moral truth that it is unjust to punish the innocent, and the humane directive to impose punishment only when it would do some good. How would one challenge the limiting retributivist position and show that there is an affirmative moral duty to prosecute and punish serious criminals? There are two possibilities. One might start from the more obvious restrictive principle, barring punishment of the innocent, and argue that this principle itself entails or implies an affirmative duty to punish the guilty. Alternatively, and as I shall argue shortly, the affirmative duty could rest on independent grounds. An argument of the first type initially seems plausible. If you agree that no one should be punished beyond what she deserves, how do you then argue that someone who does deserve punishment need not receive it?106 To say A deserves B seems to mean that A should receive B, at least in the absence of a stronger countervailing reason. But when we try to explain why this is so, to avoid relying on a merely formal concept, things become more difficult, seeming almost beside the point. According to Herbert Morris’ “free-rider” conception, a criminal deserves and must receive punishment in order to redress the unjust advantage over law-abiding citizens he acquired by his crime. The criminal benefits from the protection of the criminal law, but doesn’t suffer the burden “others have assumed, the burdens of self-restraint.”107 By forcing the defendant to suffer a theoretically-equivalent burden, punishment restores equality
2620 (noting that the rights to redress and equality are not sufficient to require punishment, which should be deemed a collective good to prevent future atrocities). 105. KANT, supra note 91, at 107 (arguing that capital punishment must be carried out even by a society about to disband, despite the lack of future utilitarian benefits for anyone). 106. Some arguable possibilities include the following: (1) a waiver theory, whereby the offender by her act consents to be subject to a kind of prosecution lottery in which all offenders of the same type have an equal risk of punishment but only some receive it; (2) a hybrid theory separating the regime of punishment, which should exist only if it produces utilitarian benefits, from its distribution of individual punishments, each of which must be limited to what is deserved; or (3) a balancing theory, as suggested by Joel Feinberg, who argued that if someone deserves certain treatment, this always constitutes a reason to do so, but one that can be overridden by other considerations. JOEL FEINBERG, Justice and Personal Desert, in DOING AND DESERVING: ESSAYS IN THE THEORY OF RESPONSIBILITY 55, 60 (1970). For variations on the second argument, see generally Rawls, supra note 56; HART, supra note 87, at 9; NINO, supra note 79, at 185. 107. HERBERT MORRIS, ON GUILT AND INNOCENCE: ESSAYS IN LEGAL PHILOSOPHY AND MORAL PSYCHOLOGY 34 (1976).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

836

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

between the criminal and the law-abiding, and the proper balance of benefits and burdens that attach to each.108 Another theory says that one must prosecute and punish a guilty defendant in order to respect him as a responsible moral agent.109 Only if we both punish the guilty and absolve the innocent can verdicts and sentences reflect the defendant’s own moral choices.110 Proponents of this view sometimes speak of the right of every individual to be respected as a responsible moral agent, and of retributive justice as therefore reflecting a right of the criminal to be punished.111 A third theory also posits that the defendant has a moral right to punishment, but traces it to a defendant’s right to atone for his crime and thereby restore his standing as an equal member of the community.112 These theories are all more or less defendant-centered, setting the appropriate sentence according to a single moral yardstick—the defendant’s desert. The restrictive principle forbids punishing the innocent; the affirmative duty requires proportional punishment of the guilty. Justice to the victim effectively gets treated as if it were a mere by-product of, or necessarily identical to, the punishment that is required to treat the offender as she deserves. Yet, it should be obvious that in the aftermath of a crime, justice is due to both the victim and the accused. This seems undeniable, and it clearly raises a
108. This is a kind of corrective justice, but one well-noted problem with the theory is that no characterization of this burden seems able to explain why crimes like rape and murder should receive more severe punishments than larceny. If the self-restraint exercised by the law-abiding consists in obedience to law, all crimes should be punished to the same degree; and if it consists in refraining from the particular crime of rape or murder, for most people that isn’t a burden at all. For additional discussion on that point, see David Dolinko, Some Thoughts about Retributivism, 101 ETHICS 537, 545–46 (1991); Jean Hampton, The Retributive Idea, in FORGIVENESS AND MERCY 111, 114 (Jeffrie G. Murphy & Jean Hampton eds., 1988). 109. See, e.g., MORRIS, supra note 107, at 46–49. 110. Id. In Kantian terms, the defendant has willed his own punishment by his choice; and only by inflicting it can we respect him as an end in himself. So Kant has said: “If you slander another, you slander yourself; if you steal from another, you steal from yourself; if you strike another, you strike yourself; if you kill another, you kill yourself.” IMMANUEL KANT, THE PHILOSOPHY OF LAW 195 (W. Hastie trans., Clark 1887). In Kant’s conception, the defendant’s freely chosen action must be interpreted as willing a universal maxim, so that as the defendant treats another, so should all (including the defendant) be treated. Id. 111. G. W. F. HEGEL, ELEMENTS OF THE PHILOSOPHY OF RIGHT 126, § 100 (Allen W. Wood ed., H. B. Nisbet trans., Cambridge Univ. Press 1991) (“The injury . . . is also a right for the criminal himself, that is, a right posited in his existent will, in his action.”); see also Markus Dirk Dubber, The Right to Be Punished: Autonomy and Its Demise in Modern Penal Thought, 16 LAW & HIST. REV. 113, 115–16 (1998). But see Quinton, supra note 104, at 57– 58 (arguing that it is “an odd sort of right whose holders would strenuously resist its recognition”); John Deigh, On the Right to be Punished: Some Doubts, 94 ETHICS 191 (1984). 112. See PACKER, supra note 104, at 38–39.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

837

question that is seldom considered: What is the relation between the two? It may not be the case that justice to the victim consists in nothing more and nothing less than insuring that the defendant receives her just deserts. Consider an alternative conception of retributive justice that does treat the victim as an independent subject of justice (and expands the options available to the ICC, as argued in Part III). This conception is best illustrated by a recent prosecution that was initiated for reasons of retributive justice rather than any utilitarian considerations. The defendant, a former Kleagle of the Ku Klux Klan named Edgar Ray Killen, was tried and convicted in June 2005 for orchestrating the notorious killings of the three young civil rights workers James Chaney, Andrew Goodman, and Michael Schwerner, that occurred in Mississippi in 1964.113 The state of Mississippi had chosen to bring no charges against anyone for all those years, despite clear evidence implicating Killen and others.114 Then in 2005, four decades after the victims were discovered buried in an earthen dam, with Killen now a frail eighty year old man in a wheelchair, the state finally brought murder charges that ended in a manslaughter conviction and a sixty year jail sentence.115 Many believe that this trial, and similar cases that rectified years of impunity, are very important morally, but why? Why is doing nothing an unacceptable injustice? The essential reason to punish Killen is neither a free rider theory that would have the defendant disgorge the “unjust advantage” acquired through his murders, nor that he effectively consented to punishment by his acts, although they are both reasons the defendant cannot properly complain about his punishment. These are prerequisites that make it permissible to punish the criminal, but they are not reasons to do it. The essential reason Mississippi has an affirmative duty to hold Killen to account is that otherwise it reinforces, and becomes complicit in, the crime and its debasement of the victims. We can see that clearly in this case, because that is what happened. Killen treated the victims as if they were mere objects, of no intrinsic worth, and by its response the government could either ratify or repudiate their subjugation. Mississippi chose to ratify it, to humiliate and devalue the victims by ignoring the killings for forty years, and

113. Emily Wagster Pettus, Ex-Klansman Gets 60 Years in Slayings, BOSTON GLOBE, June 24, 2005, at A2, available at http://www.boston.com/news/nation/articles/2005/06/24/ ex_klansman_gets_60_years_in_slayings. 114. Id. 115. Id.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

838

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

righting that wrong was what made the trial morally imperative.116 When the wrong was finally righted with the manslaughter conviction, James Chaney’s brother Ben described the reaction of their eighty-two year old mother: “She finally believes that the life of her son has some value to the people in this community.”117 Trying Edgar Ray Killen was an official act of solidarity, however belated, with all those who had been victimized by these and similar events. The Killen case is a good example of one type of case that may dominate the ICC’s caseload: crimes that are designed to devalue a whole race or people, or in the case of genocide, designed to destroy the group entirely. It also, I believe, points to an alternative conception of retributive justice—call it victim-conscious retributivism—which I want to explore in some detail. On this retributivist conception, the restrictive principle and affirmative duty reflect distinct moral concerns. Like all retributive theories, its restrictive principle embodies an obligation of justice to the defendant, protecting her from punishment beyond her desert; but on this alternative conception the affirmative duty primarily involves justice to the victim. We know that there is a duty of justice to the victim because, at a minimum, we know this much: A society treats a victim unjustly in the aftermath of crime when it does nothing, when it ignores the fact of victimization, leaving her an outcast and unrecognized as a part of the social compact. Kant, and such contemporary theorists such as Jean Hampton and George Fletcher, diagnosed a society’s failure to sanction crime as a kind of complicity in it: “the victim’s blood is on our hands,” argued Fletcher.118 Retributive justice must include an affirmative duty because of the injustice done to victims when nothing is done.119
116. As the District Attorney said to the jury in closing argument, “You can either change the history that Edgar Ray Killen and his friends wrote for us or you can confirm it.” Brian MacQuarrie, Jury Deadlocked in Miss. Trial, BOSTON GLOBE, June 21, 2005, at A2. 117. Shaila Dewan, Ex-Klansman Guilty of Manslaughter in 1964 Deaths, N.Y. TIMES, June 22, 2005, at A1. 118. GEORGE FLETCHER, WITH JUSTICE FOR SOME 6, 203 (1995); see also Hampton, supra note 90, at 1684–85, 1692; Hampton, supra note 108, at 131; Richard Burgh, Guilt, Punishment, and Desert, in RESPONSIBILITY, CHARACTER, AND THE EMOTIONS 316, 330 n.30 (Ferdinand Schoeman, ed., 1987).. George Fletcher and Jean Hampton are most explicit among a small number of contemporary theorists who view prosecution and punishment as a duty of justice that is owed to the victim. These theories may be taken as an instance of the “expressive” or “denunciatory” theory of punishment, originally formulated by Lord Denning. 119. What must be done will depend in part on the crime, because maliciously premeditated murders necessarily attach a value of zero to the victim, while manslaughters and lesser crimes do not. For a detailed explanation of why this is so, see Hampton, supra

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

839

This duty involves recognizing and repudiating victimization, and reaffirming the victim’s equal value and standing in the society. One can conceive of this obligation as a state duty of solidarity, which is correlative to a solidarity right of each citizen. Certainly, criminal punishment can be a means of fulfilling this duty, but that is not to say that punishing the criminal is the affirmative retributive obligation. Some people would question whether this victim-conscious conception should be understood as a species of retribution, or should be categorized as a different idea of punishment entirely.120 At the least, it should be recognized as no less a species of deontological justice than the strict retributivist version—it is a duty on the authorities to act justly, without calculating the results. In the Killen case the three murder victims were dead, so they themselves would obtain no benefit from the trial or the solidarity it expressed, but no one would say that the obligations of justice were extinguished along with the victims.121 Nor does the retributive duty vary according to
note 90, at 1677–82. 120. See, e.g., Quinton, supra note 104, at 56. Quinton sees the idea that punishment is required to “annul the crime” as utilitarian, not retributive, “[f]or it holds that the function of punishment is to bring about a state of affairs in which it is as if the wrongful act had never happened. This is to justify punishment by its effects, by the desirable future consequences which it brings about.” Id. This seems to me to be taking a metaphor too far. Of course the crime can never be undone, but it can be repudiated, and this is what justice demands without regard to some further outcome. Part of the confusion is that a theory that justifies punishment as a communicative act—a so-called denunciatory or expressivist theory—may do so on utilitarian grounds, or deontological grounds, or both. It may view the condemnatory message punishment generates either as a means to social betterment (see, e.g., EMILE DURKHEIM, THE DIVISION OF LABOR IN SOCIETY 108 (G. Simpson, trans., 1964)), or as an intrinsic good, required by justice to vindicate the law and reaffirm the victim’s right (see, e.g., Igor Primoratz, The Middle Way in the Philosophy of Punishment, in ISSUES IN CONTEMPORARY LEGAL PHILOSOPHY: THE INFLUENCE OF H. L. A. HART 193 (R. Gavison, ed., 1987)). If there is a problem with accepting the victim-conscious theory as a species of retributive justice in particular, it is not because the theory is necessarily and exclusively utilitarian. It is rather a problem of usage. In common parlance, retribution is often understood as a moral imperative that the defendant be punished so that she receives her just deserts. On the other hand, Jean Hampton, who developed a victim-centered theory of criminal justice, characterized her theory as retributive. See Hampton, supra note 90, at 1659. Hegel’s retributivism derived from the imperative of justice to “annul the crime” and “restore the right” of the victim. David E. Cooper, Hegel’s Theory of Punishment, in HEGEL’S POLITICAL PHILOSOPHY: PROBLEMS AND PERSPECTIVES 151 (Z. A. Peleynski, ed., 1971). And the retributivist theory espoused by Kant could arguably accommodate the victim-conscious approach, because his concern was to correct the imbalance between defendant and victim by the only method he thought could restore equality—lex talionis, an eye for an eye. 121. It is true, of course, that the victims’ families and friends, and society generally, may benefit in various ways: Some degree of psychological satisfaction and security, norm reinforcement, empowerment of a still-subjugated race, deterrence, etc., may result.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

840

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

the needs or desires of surviving victims, or afford them rights to shape the remedy; its point is to reaffirm the equal moral standing that the victim shares in common with all others.122 In other retributivist theories, the duty is to bring the criminal down by imposing suffering or deprivation. In the victim-conscious theory, the societal duty is to recognize victimization and reaffirm the equal worth of all against the predations of others. If the state’s retributive duty is to recognize victimization and reaffirm the value and rights of the victim, then arguably we should understand criminal punishment as simply a means of fulfilling this duty, and be open to the possibility that in some situations, non-penal instruments may be better able to do so. Could this ever be the case? Some might say that there are no such cases—that the only way to rectify the debasement of the victim is by inflicting the appropriate punishment. But there are reasons to doubt this. One is that no acceptable version of retributive justice can specify what particular punishment is appropriate. The one version that does claim to identify appropriate punishments is universally rejected as barbaric. This is the lex talionis, which mandates that the criminal receive as punishment the precise suffering he inflicted on the victim: “an eye for an eye.” But no one regards raping a rapist, or torturing a torturer, as appropriate punishment today. Retributivists instead demand proportional punishment, so that the gravest crimes receive the gravest punishments—which in most countries means the longest term of imprisonment. But this cannot tell us anything about where the schedule of imprisonments should begin or end, or how large the range should be; we must instead try to give retributive sense to a system of rankings with no “anchoring points.” Moreover, a second
Retributive justice is likely to produce utilitarian gains, but the retributive duty does not depend on them or any particular outcome. 122. In this respect it differs from what is commonly described as restorative justice, although it is compatible with it. Victim-conscious retributive justice is a form of objective justice. By contrast, restorative justice mechanisms may approach mediation and develop resolutions that are molded by the parties. In contemporary descriptions, restorative justice permits the reaffirmation of the victim’s worth to come from the offender rather than the state, is centered on the specific victim and his needs rather than a response required by justice to victimization, and in some renditions, permits restorative measures that are unconstrained by the defendant’s degree of desert. Llewellyn, supra note 44, at 103 (describing restorative justice as “a process whereby all the parties with a stake in a particular offence come together to resolve collectively how to deal with the aftermath of the offence and its implications for the future”); Leena Kurki, Restorative and Community Justice in the United States, in 27 CRIME AND JUSTICE: A REVIEW OF RESEARCH 235, 236 (Michael Tonry ed., 2000) (stating that core restorative justice ideals “imply that government should surrender its monopoly over responses to crime to those who are directly affected— the victim, the offender, and the community”).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

841

proportionality problem afflicts punishments for crimes against humanity and genocide. For these crimes, no penalty schedule is truly consistent with the proportionality principle, because the harshest punishment—life imprisonment, or in some countries the death penalty—is already applied to far lesser crimes.123 Even the lex talionis cannot generate a fit punishment for genocide. None of this refutes the claim that criminal punishment of some kind is required to vindicate the victim, even if the particular sentence is necessarily dictated by social, economic, and historical expectations peculiar to a particular culture. But it does suggest the possibility that criminal punishment itself is an historical and contingent means of fulfilling the retributive obligation, and not the only moral response. No doubt the state’s obligation is to repudiate the criminal act, and that repudiation must be attached to the perpetrator for it to be understood as repudiation at all. But it is not obvious that conviction and incarceration is always required to do this. Rafael Gumucio, for example, a victim of the Pinochet dictatorship, found justice in Chile’s torture report, which “in all its graphic reality, [was] a way of telling us . . . that we are still alive, and that we have a right to continue living.”124 South Africa’s TRC was intended to achieve, by means of testimony and reparations, “the restoration of the human and civil dignity of victims of gross human rights violations.”125 We can gain some insight on this issue by comparing the TRC to a regime of punishment. I take up this task in Part III, which assesses the TRC as an instrument of justice. 2. Do Obligations of Justice Override All Else?

Suppose we reject the preceding argument, and believe that the retributive duty in a particular case requires prosecution and punishment. This still does not get us to an obligation to prosecute
123. A further proportionality problem emerged in the criminal prosecutions following the Rwandan genocide, and may recur in ICC prosecutions. Rwandan defendants were divided between Rwandan courts and the ICTR in Tanzania, with those most responsible extradited to the ICTR. The proportionality problem resulted from the different sentencing powers of the two systems: The Rwandan courts were empowered to impose death sentences, while the ICTR was limited to sentences of imprisonment. The upshot was that leaders of the genocide received lesser sentences than their followers. See Mark A. Drumbl, Collective Violence and Individual Punishment: The Criminality of Mass Atrocity, 99 NW. U. L. REV. 539, 579 (2005). 124. See Rafael Gumucio, My Tortured Inheritance, N.Y. TIMES, Dec. 13, 2004, at A1. 125. 1 S. AFR. TRUTH & RECONCILIATION COMM', FINAL REPORT OF THE TRUTH AND N RECONCILIATION COMMISSION ch. 4, ¶ 32(c) (1998), available at http://www.stanford.edu/ class/history48q/Documents/EMBARGO/1chap4.htm [hereinafter TRC FINAL REPORT].

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

842

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

and punish at any price. Could any duty be so important that we are bound to disregard the consequences of acting on it, however ruinous they may be? As Samuel Scheffler has noted, the idea runs against a natural and familiar conception of rationality as instrumental and goalmaximizing;126 but most people believe that good ends cannot justify some means. Hence the outcry against torture at Abu Ghraib, despite any potential it may have had for uncovering terrorist plans. The best example of such a proscription is the transplant hypothetical; everyone agrees that it is immoral for a doctor to kill a person in order to distribute his organs to five desperate transplant candidates, even though her act of killing would save five lives at the cost of one. Promises should be kept, lies foresworn, and rights observed—these putative constraints on maximizing utility all illustrate a deontological position that some acts are mandatory (or forbidden) regardless of how much harm (or good) they might produce.127 Is retributive justice like that, so that deserved punishment must be inflicted even at great cost to innocents? Kant argued that case in the eighteenth century,128 and many observers of the ICC argue that case today, including an NGO named “No Peace Without Justice.” For deontologists, the issue becomes especially complicated when treating retributive justice as an absolute requirement would destroy other moral rights they regard as equally absolute. If ICC charges against LRA leaders would prolong the war as the Acholi leaders claim, the result would be not simply deaths, but killings perpetrated in violation of a non-negotiable right to life. Should the ICC not indict, thereby perpetrating an injustice in order to avoid a greater injustice? War crimes prosecutions are prone to such dilemmas. Sometimes prosecuting some individuals would result in many more escaping punishment.129 Sometimes obtaining
126. Samuel Scheffler, Introduction, CONSEQUENTIALISM AND ITS CRITICS 1, 9 (Samuel Scheffler ed., 1988). 127. See, e.g., John Rawls, Justice as Fairness, 67 PHILOSOPHICAL REV. 164, 194 (1958) (arguing that maximizing utility is limited to means that are fair); ROBERT NOZICK, ANARCHY, STATE, AND UTOPIA 28–33 (1974) (describing “side constraints” on maximization); Thomas Nagel, Autonomy and Deontology, in CONSEQUENTIALISM AND ITS CRITICS 142, 143–44 (Samuel Scheffler ed., 1988). 128. Kant argued that even if someone sentenced to death volunteered to be a subject of medical experimentation that would result in great benefits, the death sentence should be carried out. In his view, retribution is a non-negotiable obligation. KANT, supra note 91, at 105. 129. In Argentina, for example, prosecutors of the military junta calculated how many prosecutions they could pursue without threatening the entire project. See supra note 97. This problem routinely confronts domestic prosecutions when providing immunity to one defendant is necessary to obtain the testimony necessary to convict others. In the case of the

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

843

indictments may induce a criminal regime to cling to power, leaving that country’s population consigned to suffer continued violations of their most fundamental human rights. Dworkin famously described such rights as “trumps,” because they demarcate areas in which the individual is inviolable, and thus trump even a majority’s competing preferences or interests.130 These rights and moral duties are not supposed to be subject to trade-offs—but in the above cases, observing one non-negotiable right or duty will sacrifice another. Each cannot trump the other, so how should one assess this collision of absolutist moral demands? Embracing a moral imperative to bring criminals to justice does not answer this question, unless accompanied by reasons why this moral command should have priority over the others. Sometimes it doesn’t have priority. Even strict retributivists acknowledge some side constraints on retributive justice. This is why those retributivists who in principle accept the rule of lex talionis—an eye for an eye—place limits on that rule. Retributivists such as Jeffrey Reiman and Jeffrie Murphy reject punishments of rape for a rapist, and torture for a torturer, not because the torturer does not deserve torture, but because of a side constraint on imposing punishments which would violate human dignity. They adhere to the strictest interpretation of the retributivist proportionality requirement, but do not deem it absolute—there is another moral consideration that comes first.131 If authorities may not treat a convicted defendant in certain ways because they violate human dignity, should not they also be barred from pursuing prosecutions that would lead to the same kinds of mistreatment of innocent third parties by prolonging a war? Here are three ways we can think about this question. First, we can view duties of justice as imposing an absolute, first-person obligation on the ICC—an obligation that the Court is flatly forbidden to violate in its own conduct, however much good that would do. This is a deontological purist’s view, insisting that we must conform to the same moral rules regardless of the particular context and costs involved. Second, we might take an aggregative approach, and define the ICC’s obligation as the minimization of injustice overall,
ICC, which is dependent on state cooperation to bring its cases, it is easy to envision a troubling variation—a government that attempts to protect its own offenders as a cost of access to others. 130. See RONALD DWORKIN, TAKING RIGHTS SERIOUSLY xi, 231–38 (1977); Ronald Dworkin, Rights as Trumps, in THEORIES OF RIGHTS 153–67 (Jeremy Waldron ed., 1984). 131. See JEFFRIE MURPHY, RETRIBUTION, JUSTICE AND THERAPY 243 (1979); Murphy, supra note 95, at 107; see generally Jeffrey Reiman, Justice, Civilization and the Death Penalty, 14 J. PHIL. & PUB. AFF. 115 (1985).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

844

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

even if it must ignore a warranted case or perpetrate some other injustice to do so. The third way emerges as an intuitive response to the other two, both of which posit rigid principles that can lead to fanatical applications. It says the Court is obligated to adhere to the obligations of retributive justice, but also incorporates the common intuition that there are limits to these obligations in catastrophic circumstances. That is the non-absolutist position that I believe the ICC should adopt, largely because the other two defy common moral sense. This is what I will now try to show.
a. The Deontological Purist’s View: Clean Hands Above All

Deontological duties are concerned what we must or must not do—with right and wrong conduct, not its results. Some people believe that these duties are absolute,132 but it is important to understand the limited sense of this claim. It cannot mean that a person must always adhere to every deontological duty, because one duty may conflict with another in a particular situation. The claim is rather that in choosing what to do, one must give deontological obligations absolute priority over all other considerations. Because these obligations apply only to one’s own acts, for a deontological absolutist they necessarily override concerns regarding not only the outcomes of these acts, but also the unjust acts others may commit in response to one’s own.133 Consider how this distinction applies to a deontological imperative to bring criminals to justice. Jeffrey Reiman and Jeffrie Murphy say that this retributivist imperative has its limits, but these limits exist only because state is also bound by a second deontological constraint on its own actions—an obligation not to impose cruel and degrading punishment.134 No such additional, conflicting deontological constraint exists in the ICC’s “peace v. justice” scenario, however. If ICC prosecutions prolong a war, the resultant killings are committed by others, and the Court has violated no deontological duty of its own; it is responsible for living within the rules itself, but not responsible for whether others do.135 The
132. See, e.g., CHARLES FRIED, RIGHT AND WRONG 7, 9 (1978); Michael S. Moore, Torture and the Balance of Evils, 23 ISR. L. REV. 280, 328 (1989) (attributing the “whatever the consequences” aspect of deontology to Kant). 133. For detailed exploration of this point, see THOMAS NAGEL, THE VIEW FROM NOWHERE 175–85 (1986). 134. See supra note 131. 135. See Bernard Williams, A Critique of Utilitarianism, in UTILITARIANISM: FOR AND AGAINST 77, 89 (J. C. C. Smart & Bernard Williams eds., 1973).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

845

problem, of course, is that sometimes doing justice will lead others to respond unjustly. Bringing a prosecution may result in defendants killing witnesses, for example. But orthodox deontology still draws the same line: We are not permitted to do a little evil in order to prevent someone else from doing more. Without going into detail, the special and absolute responsibility we have for our own acts directly reflects the deontological conception of individual autonomy in many ways. For example, the same assumption of free will that allows us to be held morally responsible for our own acts applies to limit our responsibility for another volitional agent’s acts that ensue from ours.136 There is a second distinction between the two cases that reinforces the first: The ICC prosecutor’s intention is to deliver justice, and the ensuing killings are an unsought side-effect. But when the state imposes degrading punishment, its intent is to do just that. There is a moral chasm between aiming to do harm, and acting for noble reasons in spite of the harmful side-effects. (Compare the hostage taker who deliberately kills his captive with the government that refuses her demands to avoid encouraging future hostagetaking.137) The upshot of both distinctions combined is that, on the orthodox account, one’s deontological obligations are absolute, but they are also absolutely confined to one’s own deliberate acts. The central problem with these rigid rules is, of course, their absolutism: They take what seem to be important moral considerations entirely off the table.138 If we take there to be a
136. On this highly contentious question, see Sanford H. Kadish, Complicity, Cause, and Blame: A Study in the Interpretation of Doctrine, 73 CAL. L. REV. 323, 327 (1985) (arguing that the “conception of causation appropriate to physical events is out of place in the human realm”). 137. This point also explains that part of the law of war which distinguishes between terrorist bombings that deliberately target a civilian population and bombings aimed at military targets that will foreseeably kill civilians. For elaborations on the moral distinction between intended and foreseeable harms, see Thomas Nagel, War and Massacre, in CONSEQUENTIALISM AND ITS CRITICS 51, 58 (Samuel Scheffler ed., 1988) (noting the parallel Catholic theological doctrine of double effect); see also Vacco v. Quill, 521 U.S. 793, 808– 09 (1997) (finding no violation of equal protection in a ban on assisted suicide because there is a morally relevant distinction between intending the death of a terminally ill patient and foreseeing death as a result of palliative care). But see H. L. A. HART, Intention and Punishment, in PUNISHMENT AND RESPONSIBILITY: ESSAYS IN THE PHILOSOPHY OF LAW 123, 125 (1968) (entirely rejecting the distinction between intended and unintended but foreseen results as “the result of a legalistic conception of morality”). 138. It is also worth mentioning another difficulty: These moral principles are quite different from those the ICC would itself apply to a defendant. The mens rea doctrine does not provide a general defense to those whose well-motivated acts produced foreseeable but undesired results, as the deontological intent limitation does. Governing doctrine also provides a choice of evils defense, justifying otherwise criminal conduct when necessary to

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

846

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

retributive duty to bring criminals to justice, they posit that the ICC has an absolute duty to assure prosecution, rendering irrelevant any undesired destructive impact it would have. It does not matter if that impact is catastrophic, or fully foreseeable, or wholly set in motion by the prosecution. Kant believed this, and is famous for a parallel deontological claim—that one is obligated not to lie, even when it means divulging the whereabouts of someone to an enemy bent on killing him.139 Most people would no doubt regard this as the view of a fanatic, or of someone lacking basic humanity. The challenge is how to properly respond to such dangers yet still recognize the obligations of justice in a principled way. Consider two alternatives to the purist view, each of which treats obligations of justice as a priority, but one that can be overridden by other moral considerations.
b. The Minimize Injustice View

On the orthodox deontological view, the ICC is obligated to keep its hands clean, to act justly at all times. If retributive justice requires a prosecution, it must proceed regardless of cost. One alternative is to say that the true moral obligation of a state, and the ICC in its stead, is to minimize injustice overall, even if it must itself perpetrate an injustice to do so. On this principle, many amnesties for peace may well have been morally required, on the plausible assumption that the denial of redress inflicts less injustice than the prolonged slaughter of many war victims. One should distinguish two separate issues that may be involved in the minimize injustice calculation. One concerns the strength of the deontological imperative at issue. Many may consider the denial of retributive justice to be a weak deontological violation, not comparable to the evils of killing and other human rights violations. Few deontologists are likely to be absolutists with regard to the retributive justice imperative; they are more likely to apply the kind of “threshold deontology” analysis, allowing for consequentialist overrides, that I describe in the next section. The other issue concerns the number of violations that would ensue from one’s choice of options, and it is here that the minimize injustice principle is most significant and troublesome.
avert a greater harm from occurring. 139. See Immanuel Kant, On a Supposed Right to Tell Lies from Benevolent Motives, in KANT’S CRITIQUE OF PRACTICAL REASON AND OTHER WORKS ON THE THEORY OF ETHICS 361–65 (T. K. Abbott trans., Longmans, Green, and Co. 1909).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

847

To explore the problem of doing one injustice to prevent someone else from perpetrating more, let us consider a simplified example in which the kind of injustices involved on both sides of the ledger are the same—killings. Suppose you believe that every state execution is a wrongful killing, and no less unjust than a street murder. If you are a deontological purist, you will be a death penalty abolitionist under all circumstances. But if you are an injusticeminimizer, you will believe that the state must institute capital punishment if doing so would deter more wrongful killings than it inflicts. Recently Professors Sunstein and Vermuele have argued precisely this position.140 They are willing to grant that state executions may be no less wrongful than private murders,141 but argue that because the state is responsible for private killings it fails to prevent as well as killings it directly perpetrates, the morality of capital punishment reduces entirely to a question of numbers: Will state killings reduce or increase the number of killings overall?142 In their view, governing officials are morally obligated to undertake such life-life tradeoffs. There are a number of logical and other problems with the Sunstein-Vermeule argument that I cannot explore here.143 But I do want to comment briefly on their claim that governing authorities have a duty to perpetrate injustice whenever it would minimize injustice overall—a claim that, if true, might be thought to apply to the ICC as an international governmental organization that assumes a state’s obligations of justice only upon the latter’s default. My counterclaim is that the aggregation this requires is itself a form of injustice that necessarily goes uncounted in their minimize-injustice calculation. To see this, ask yourself whether you would prefer to live under a government that practiced the life-life tradeoffs the authors advocate, solely by virtue of the increase in your odds of survival.
140. See Cass R. Sunstein & Adrian Vermeule, Is Capital Punishment Morally Required? Acts, Omissions, and Life-Life Tradeoffs, 58 STAN. L. REV. 703 (2005). 141. Id. at 707–08, 718–19. Sunstein and Vermeule argue that: [A]ny deontological injunction against the wrongful infliction of death turns out to be indeterminate on the moral status of capital punishment if the death is necessary to prevent significant numbers of killings . . . . By itself, the act of execution may be a wrong . . . . But the existence of life-life tradeoffs raises the possibility that for those who oppose killing, a rejection of capital punishment is not necessarily mandated. On the contrary, it may well be morally compelled. Id. at 707–08 (emphasis added). 142. Id. at 705. 143. I will explore these problems in detail in an article to follow, entitled Killing in Good Conscience: Comments on the Sunstein and Vermeule Lesser Evil Argument for Capital Punishment and other Human Rights Violations.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

848

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

You would be wise not to, because in exchange for a minute improvement in these odds, you would be giving up something of great value–the state’s recognition of your personhood, or in different words, of your status as an inviolable and intrinsically valuable being. This legal status is of fundamental significance for every person, separate and apart from whether that person is victimized or not. Thomas Nagel says, “to be tortured would be terrible; but to be tortured and also to be someone it was not wrong to torture would be even worse.”144 Legal inviolability is the state’s recognition of the intrinsic value of each human life, which generates duties of justice to each person individually.145 On the minimize injustice view, individuals have a very different legal status, that of instruments that may be used for the collective benefit. If torturing some would reduce torture overall, or executing drunk drivers would deter some greater number of fatal collisions, the state is obligated to do so. The idea suggests a powerful public sector with the authority to distribute and redistribute the most extreme deprivations so as to achieve the lowest total number of such deprivations. Citizens cease to be rightsholders; instead, the state acquires the right to decide some shall die so more can live. The minimize-injustice calculation takes no account of this injustice, which arises when the state in particular undertakes such life-life tradeoffs. There is an obvious kinship between this objection and the view expressed earlier concerning the nature of retributive justice. Both are premised on the human rights axiom that every individual possesses a moral status of inviolability that the state and its laws must respect. The State of Mississippi did not do so following the murders of Goodman, Chaney, and Schwerner for forty years, but treated them as having no moral status at all.146 The minimizeinjustice principle does so as well, but wholesale, as it strips the legal recognition of this status from every citizen at once. In both cases, authorities have a special responsibility to act justly, because legal inviolability is a status that governments and courts but not individuals can destroy.
144. Thomas Nagel, Personal Rights and Public Space, 24 P HIL . & P UB . AFF. 83, 93 (1995), citing Frances Myrna Kamm, Harming Some to Save Others, 57 PHIL. STUD. 251 (1989); Frances Myrna Kamm, Non-consequentialism, the Person as an End-in-Itself, and the Significance of Status, 21 PHIL. & PUB. AFF. 381 (1992). 145. On a human rights conception, moral inviolability is a moral status that cannot be taken away so long as the person exists. The state should reflect this moral right by enacting certain kinds of legal inviolability. Unlike moral inviolability, legal inviolability is a status that can be repealed by the state or violated by others. 146. See supra notes 114–18 and accompanying text.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

849

c.

The Catastrophic Exception View: “Threshold Deontology”

Both of the above theories quickly lead to extreme applications that very few would accept. A white lie about someone’s whereabouts is surely required when telling the truth would facilitate his murder, contra Kant’s deontological absolutism. And the minimize-injustice view, which licenses the authorities to unjustly kill one so that two putative murder victims survive, forsakes human rights in the name of rights. Here is a third possibility which does not force us to such unacceptable conclusions. This view does recognize the inviolable status of individuals, but it also incorporates the common intuition that this status has a limit. That limit is the catastrophic exception. Put in different words, a person’s right to justice, and the ICC’s duty to deliver it, are of great weight, but they are not absolute and can be overridden in extreme circumstances. This position is often labeled threshold deontology—the threshold is that level of aggregate harm (or, alternatively, aggregate injustice) at which deontological rights and duties are overridden.147 Below that threshold, a person’s human rights cannot be impinged, even if doing so would produce great benefits or avert serious harms to others. But violating someone’s rights is justified when the aggregate harm that would result reaches the catastrophic threshold.148 This view gets much of its force from the intuitive responses most people have to extreme scenarios such as the ticking bomb hypothetical, where torturing a terrorist’s child is the only way to reveal the location of a nuclear bomb in time to defuse it. That intuition distills into the view that, in the words of Michael Moore, “a very high threshold of bad consequences . . . must be threatened before something as awful as torturing an innocent person can be justified.”149 Some theorists argue that the very conception of a threshold is incoherent.150 A recent article suggests that the permissibility of coercive interrogations should be judged according to a straight costbenefit analysis, without requiring a catastrophic threshold to be reached.151 The authors find the latter idea simply arbitrary. “What
147. See Moore, supra note 132, at 330. 148. Id. For a thorough but critical explication of threshold deontology, see Larry

Alexander, Deontology at the Threshold, 37 SAN DIEGO L. REV. 893 (2000). 149. Moore, supra note 132, at 330. 150. See generally Alexander, supra note 148; see also Posner & Vermeule, supra note 74, at 677 n. 12. 151. Posner & Vermeule, supra note 74, at 677.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

850

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

is quite mysterious,” they write, “is why the sheer catastrophic size of the threatened harm should matter. The obvious alternative is to say that the harm prevented must simply be greater than the harm inflicted.”152 But this “obvious alternative” guts human rights of their meaning, as we have just seen in our discussion of a minimizeinjustice rule. Only the threshold conception instantiates the common intuition that at some point grave harm should count, but in a way that does not destroy the respect for an individual’s intrinsic value that rights provide. This idea is no more mysterious than the idea that a value can be vitally important without being absolute. It is commonplace in American constitutional interpretation, of course— fundamental rights constrain, but are defeasible by a “compelling state interest.” The same idea structures the European Convention on Human Rights, which provides that many of its enumerated rights may be overridden in some circumstances where the collective interest is at stake, and which specifically details the few such circumstances in which the right to life may be abrogated.153 What these doctrines capture is the idea that an individual’s inviolability with regard to vital interests is of great, but not infinite, moral weight. The threshold conception is not without its difficulties, but they are difficulties that extend to many sorts of value judgments we nevertheless find credible and indispensable. For example, one prominent complaint is that any precise specification of the threshold will be arbitrary. Why do rights get overridden at N rather than N-1? Line-drawing commonly provokes such complaints, but they do not undermine the qualitative difference that the line imperfectly represents. If I put sugar in my coffee, no particular granule changes the coffee from bitter to sweet, but it would be foolish to deny that there is a difference between the two.
152. Id. (emphasis added). Some deontologists are equally skeptical. For example, Larry Alexander notes that at the threshold, we are supposed to “switch from not being resources for others to being resources for others,” and argues that “it seems downright implausible that the moral universe is so constituted. There may be thresholds at which new phenomena emerge, but it is quite another thing to have thresholds at which things become their opposites.” Alexander, supra note 148, at 912. 153. European Convention for the Protection of Human Rights and Fundamental Freedoms, as amended by Protocol No. 11, Nov. 4, 1950, 213 U.N.T.S. 221, available at http://conventions.coe.int/treaty/en/Treaties/Html/005.htm. For example, Article 9(2) of the European Convention states: Freedom to manifest one’s religion or beliefs shall be subject only to such limitations as are prescribed by law and are necessary in a democratic society in the interests of public safety, for the protection of public order, health or morals, or for the protection of the rights and freedoms of others. Id. art. 9(2). The right to life is guaranteed under Article 2.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

851

A more substantial complaint is that we have no method of determining where the threshold lies, other than one’s individual, perhaps idiosyncratic moral intuition. If I believe torture is only justified when it is the sole means of preventing thousands of deaths, and you believe torture is justified to save twenty lives, we will have almost nothing to say to each other, and no principled way of resolving the question.154 However, we often have to choose among plural, even incommensurable, values, and choices in such cases can be intelligible even though we lack both precision and theoretical guidance. W. D. Ross provided one model of this process with his conception of prima facie duties.155 According to Ross, we learn through experience that certain issues have a moral valance, and have this quality in whatever context they arise.156 Whenever we are tempted to lie, for example, we see that it has the same negative moral valence, and that this provides a reason not to do it. Our experience thereby generates a moral world of prima facie duties— prima facie because they sometimes come into conflict with each other so that we must choose which duty to follow and which to abandon.157 We have no alternative in such cases but to decide which matters more in the particular context that confronts us.158 Jonathan Dancy, an interpreter of Ross, says that this is inescapably a matter for judgment, not theory, and that the importance of rights is simply captured “by awarding rights great weight when we come to balance reasons for and against.”159 I would add that a core criterion
154. A principle eludes us because, as Larry Alexander has pointed out, the threshold marks a shift between incommensurable ways of treating an individual—as inviolable, or as fungible and subject to trade-offs for the collective good. Each idea seems to exclude the other, and neither idea incorporates any features which would indicate an endpoint for its claims. This seems to defeat any possibility of developing a common measure that would permit them to be balanced or weighed against each other. See Alexander, supra note 148, at 907, 912. Another difficulty is to determine which factors should count towards the threshold. For example, one position might include aggregate loss of welfare or other negative outcomes that rise to a sufficiently grave level. A different and more limited view would hold that only a severe toll in human rights can justify violating a human right, not a loss in aggregate welfare. There are further complications, even on this second view. It cannot be that all rights violations count equally, so that the threshold is purely a function of the number of people who have suffered the infringement of any right. When different rights are at stake, one must also consider their importance relative to each other. We would have to define and compare the importance of rights to criminal justice with the human rights at stake on the other side, as well as the predicted numbers that would suffer the deprivation of each. For more on this issue, see generally Jeremy Waldron, Rights in Conflict, 99 ETHICS 503 (1989). 155. WILLIAM. D. ROSS, THE RIGHT AND THE GOOD 18–20 (1930). 156. Id. at 32–34, 39–41. 157. Id. at 32–34, 46–47. 158. Id. at 39–41. 159. Jonathan Dancy, An Ethic of Prima Facie Duties, in A COMPANION TO ETHICS 219,

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

852

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

for this judgment is whether allowing a particular exception respects the fundamental nature of that right. On this account, aggregate harm always counts as a moral counterforce to deontological rights and duties, but becomes overriding only at a certain point, one that we must rely on judgment to discern. Judgments will differ greatly as to where this point is, but however diverse the thresholds these judgments generate, they will be more defensible than either of the alternative criteria— deontological absolutism or unremitting consequentialist aggregation. Our intuitions change at a certain catastrophic point, and because deontological rules are largely intuitive,160 the rules change as well. Even Kant suggests that the sovereign must have it in his power to suspend the law of punishment if executing a punishment would destroy the institutions of justice themselves.161 Someone who believes that it is a duty of justice to prosecute war criminals, but that this duty can be outweighed if adhering to it would also produce massive bloodshed or tyranny, will be led to a question others need not ask: Is there a third, less debilitating option for the ICC, a more “narrowly tailored” option that can secure peace or freedom while still promoting accountability as far as possible? If there is, the ICC would not be justified in tolerating impunity even when consequentialist considerations make prosecution prohibitive. Rather, it would be justified in deferring only to that alternative form of accountability. Moreover, if such an alternative could exist, the ICC has a responsibility to ensure that it does exist if it has the means
221, 226–27 (Peter Singer ed., 1997) (citing ROSS, supra note 155, at 19). A similar argument for the threshold theory is offered by the philosopher Shelly Kagan, who says that the theory is: [J]ust what we might expect from a pluralistic view, one that accords intrinsic significance to both goodness of results and harm doing . . . . Since harm doing has a significant amount of weight in its own right, it will often outweigh goodness of results . . . . But as the amount at stake in any given case increases, the importance of the second factor will presumably increase as well—and this might well yield a threshold for that constraint. SHELLY KAGAN, NORMATIVE ETHICS 80–81 (1998). 160. Despite Kant’s effort to ground their commands in a single rational principle, the best argument for them is that the utilitarian alternative is so counterintuitive. 161. KANT, supra note 91, at 110. Most prominent contemporary anti-utilitarians, such as Rawls, Fried, Moore, and Nagel, also discern a limit to their deontological claims when aggregate costs become very great. See RAWLS, supra note 88, at 244–51; JOHN RAWLS, POLITICAL LIBERALISM 355–56 (1993); CHARLES FRIED, RIGHT AND WRONG 10 (1978); Moore, supra note 132, at 328–29; Nagel, supra note 144, at 84–85; Nagel, supra note 137, at 54. For further arguments in favor of a “catastrophic exception” to deontological rules, see generally Kai Nielsen, Against Moral Conservatism, 82 ETHICS 219 (1972); Michael Walzer, Political Action: The Problem of Dirty Hands, 2 PHIL. & PUB. AFF. 160 (1973); CHARLES BLACK, Mr. Justice Black, the Supreme Court, and the Bill of Rights, in THE OCCASIONS OF JUSTICE: ESSAYS MOSTLY ON LAW 89 (1963).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

853

to do so. It should not permit itself to be cornered into a forced choice between peace without justice, or justice without peace, if its adoption of certain policies would eliminate or reduce the number of such cases. Suppose that the prosecutor deems the South African TRC model to be a morally acceptable alternative when prosecution is not feasible. There are certainly prospective policies that could motivate and facilitate its use in appropriate cases. Without going into detail, one such set of policies for handling cases when prosecution would come at a disastrous cost might include (1) refusing to defer to blanket amnesties, (2) delaying prosecution when that would be sufficient to avert disaster, (3) deferring to a properly constituted state-based TRC-type non-penal mechanism when it is available, and (4) through its decisions, enunciating a set of standards and best practices for truth commissions and conditional amnesties, against which the acceptability of a particular state institution would be assessed. By enunciating standards, the ICC or its prosecutor can help effectuate those standards, and make available the option of peace with accountability in the kinds of situations that traditionally have presented a moral dilemma without a solution. C. Pluralism

The preceding arguments would broaden the options available to the ICC when prosecuting would threaten severe consequences. The first denies that institutions such as the South African TRC necessarily compromise justice, and holds that there are times when they may be a principled societal response to crimes of collective violence. The second argues that whatever the requirements of justice, they are not absolute and therefore the prosecutor should include the political and human costs of a prosecution in his deliberation. From the ICC’s point of view, there is more involved in these claims than whether they are persuasive. Because the ICC has jurisdiction over nationals of states with very different legal and moral cultures, it is also obliged to consider how much room there is for diverse state approaches to these issues. It can attempt to move in a hegemonic direction, and has a legal argument for doing so—all state parties to the Rome Statute have consented to be bound by the Court and to relinquish whatever degree of sovereignty that implies.162 But that legal fact does not dissolve the extensive
162. This point requires two qualifications. First, the ICC has jurisdiction over the national of a state that has not ratified the Rome Treaty when she commits a crime in a state

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

854

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

diversity among states and conceptions of justice, and a global institution would be wise to respect these differences to the fullest extent compatible with its mission. This respect is what underlies the “margin of appreciation” that the European Court of Human Rights provides to its constituent states in judging their disparate human rights practices.163 So, here is a third argument supporting ICC deference to some non-penal alternatives that a country may invoke to impose accountability and reckon with its past—the pluralist argument. The moral argument for a capacious embrace of diversity may rely on process values, or a pluralist conception of value, or both. Regarding process concerns, the ICC must address two problems that afflict many global institutions. First, how can it be impartial when its constituent states have conflicting moral and legal values? Unless it can do so, the ICC will be seen as asserting not justice but power of one group over the others. Second, how can the ICC attain democratic legitimacy when it is far removed from the people it affects? The local populations with the most immediately at stake in the decision whether to prosecute have only the weakest institutional channel for influencing the decision,164 and no means to hold the ICC prosecutor accountable for it.165 Both problems should point the ICC
that has, or that consents to jurisdiction. Second, under the Rome Statute there may be uncommon cases in which the defendant is neither a national of a state party nor accused of committing the crime within the territory of a state party, because the Security Council may refer a situation to the ICC regardless of the party status of the state. See Madeline Morris, The Democratic Dilemma of the International Criminal Court, 5 BUFF. CRIM. L. REV. 591, 592–94 (2002). 163. Lawless v. Ireland, 1 Eur. Ct. H.R. (ser. B) at 408 (1960–1961); see also Franz Matscher, Methods of Interpretation of the Convention, in THE EUROPEAN SYSTEM FOR THE PROTECTION OF HUMAN RIGHTS 63, 75 (Ronald St. J. Macdonald et al. eds., 1993) (stating that “convention institutions are not entitled to dictate uniformity,” but “are obliged to respect, within certain bounds, the cultural and ideological variety, and also the legal variety, which are characteristic of Europe”). 164. See supra note 67. 165. Richard Pildes criticizes a “dark side” of international legal institutions that are free from “democratic accountability in a more dramatic way than even the independent courts of democratic states,” despite the enormously consequential discretionary and political judgments they make. Richard H. Pildes, The Dark Side of Legalism, 44 VA. J. INT’L L. 145, 161 (2003); see also Turner, supra note 37, at 20 (stating that because there is no international legislature, there is no system that could check through legislation the ICC’s legal interpretations which are at odds with statutory text, legislative intent, or majority preferences). Of course, what some critics of the democratic footing of international legal institutions may overlook is that the same democracy deficit applies when foreign populations are affected by policies of another country, whether the latter is a democracy or not. For example, no one in Iraq had a direct or indirect voice on whether their country should suffer invasion and occupation. See Paul Mahoney, Marvelous Richness of Diversity or Invidious Cultural

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

855

away from narrow, centralized directives and towards a pluralistic framework which leaves significant scope for individual states to choose their own methods of accountability. This is an analogue of the liberal response to the incompatible values and goals of the citizenry within a single nation—the construction of an impartial framework within which people may pursue their own paths without compromising another’s ability to do so as well.166 At the international level the object is to respect and accommodate collective rather than individual choices. These are aspects of procedural justice which have a claim for that reason alone, but it is also true that failing to attend to these norms can badly undermine the ICC and its mission.167 The local populations affected will often have contextual knowledge and insights effectively unavailable to international personnel, but important in gauging the appropriateness of a prosecution. Moreover, if these people perceive themselves as marginalized, they may see the international court as illegitimately imposed, or its verdicts as biased and unjust, or their own bypassed leaders and justice system as weak and useless. If so, the prospects for moral recognition and repair, and for fostering democratic and rule of law norms, will be weakened accordingly. Many believe that such factors greatly compromised the moral authority and didactic potential of the ICTR and ICTY,168 compared to such home-grown efforts as the South African TRC and the trials of tyrannical leaders who had ruled Argentina and South Korea.169 The Rome Statute’s
Relativism, 9 HUM. RTS. L.J. 1, 3 (1998) (noting that the margin of appreciation is applied by the European Court of Human Rights because judges must give due recognition to the democratic processes of member states); Landsman, supra note 50, at 86–87. 166. See, e.g., RAWLS, supra note 161, at xvi. 167. This pragmatic concern is also cited as a justification for the margin of appreciation. See, e.g., Matscher, supra note 163, at 122 (noting that the margin of appreciation has enabled the Court to avoid damaging disputes with states over the authority of the Court and Convention). 168. See, e.g., Howland & Calathes, supra note 69, at 162–63 (finding that the ICTR’s concern with independence and neutrality lead to marginalization of Rwandans and limited impact); Kritz, supra note 50, at 133 (stating that studies of ICTR show that domestic court decisions could have greater force because verdicts would be rendered by courts familiar to the local community); Turner, supra note 37, at 24–26 nn.131–46 (stating that ICTY and ICTR have been perceived as biased, foreign, and lacking in impact); Helena Cobban, Healing Rwanda: Can an International Court Deliver Justice?, 28 BOSTON REV., Dec. 2003/Jan. 2004, available at http://bostonreview.net/BR28.6/cobban.html (finding that the ICTR lacks credibility in Rwanda). 169. See, e.g., Turner, supra note 37, at 27–28 (finding reconciliation dependent on the nation conducting its own war crimes trials, and citing examples from South Korea, France, and Argentina); Kritz, supra note 78, at 21, 27–29 (comparing national to international prosecutions).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

856

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

complementarity provisions are designed to assure a decentralized system responsive to municipal norms and interests by limiting ICC jurisdiction to cases the state has failed to prosecute. But those provisions themselves will often require interpretation and decisions determining how much variation from the criminal model is permissible. Even if one puts aside the process issues of impartiality and democratic legitimacy, there remains a substantive objection to imposing a single solution on constituent states: The possibility that there is no one best solution, but rather a number of reasonable ways a country may confront its past. On this view, at least some questions of morality have a number of valid alternative answers that are incommensurable—impossible to compare with or rank against each other. For example, I can pay my daughter’s college tuition, or donate the money to a literacy campaign. Does morality dictate the correct choice, or must I choose “for no better reason than that each value is what it is,” as Isaiah Berlin wrote, “and not because it can be shown on some single scale to be higher than another”?170 On Berlin’s account, moral pluralism is an accurate rendition of the nature of value. Others believe that inescapable epistemological weaknesses necessitate a pluralist approach.171 Perhaps some moral questions are so contextual or complex that reason is too feeble to discern the best answer, or perhaps one’s unique upbringing generates biases or blinders that one can never sufficiently overcome.172 In any event, as Rawls pointed out, the brute fact of reasonable moral disagreement calls for fairness and restraint.173 Lawyers should recognize this pluralist conception in the ubiquitous term “reasonable.” To give jurors a “reasonable person” instruction is to tell them that that there is conduct which should be recognized as legitimate, even though they might not choose to act that way themselves.174
170. Isaiah Berlin, The Originality of Machiavelli, in AGAINST THE CURRENT 25, 78 (1980). On pluralism and incommensurability generally, see ISAIAH BERLIN, THE CROOKED TIMBER OF HUMANITY (1991); Isaiah Berlin, Two Concepts of Liberty, in LIBERTY: INCORPORATING FOUR ESSAYS ON LIBERTY (2002); see also 11 SOC. PHIL. & POL’Y 1–280 (1994) (collecting articles on the subject of cultural pluralism and moral knowledge); Cass R. Sunstein, Incommensurability and Valuation in Law, 92 MICH. L. REV. 779 (1994); STUART HAMPSHIRE, MORALITY AND CONFLICT 12 (1983). 171. See, e.g., JOHN GRAY, TWO FACES OF LIBERALISM 3 (2000). 172. RAWLS, supra note 161, at 56–58. 173. Id. at xv–xvi, xxv–vi. 174. On the subject of hypothetical justifications, see J. M. Balkin, Understanding Legal Understanding: The Legal Subject and the Problem of Legal Coherence, 103 YALE L.J. 105, 117–21 (1993).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

857

A pluralist philosophy is a tolerant philosophy, but it does not validate any and all conduct, any more than the reasonable person standard accepts all conduct as reasonable. There may be a multiplicity of right moral choices, but also some choices that are unconditionally wrong.175 This fundamentally distinguishes moral pluralism from cultural relativism. Cultural relativists view cultures as black holes whose morality is inaccessible to those outside them; pluralists say we can and sometimes must make moral judgments across diversity. So pluralists stake out a normative space that neither insists that everyone follow one path nor paints a picture of self-enclosed ethical worlds which can only bump into one another. The pluralist’s quintessential moral task is to determine where reasonable diversity ends and moral imperatives begin. In this spirit, the former Vice Chairperson of the South African TRC, Alex Boraine, has urged the ICC to make “every effort . . . to assist countries to find their own solutions provided that there is no blatant disregard of fundamental human rights.”176 Reasonable moral disagreement arguably exists with regard to the problems discussed above. In the aftermath of atrocities, is a full accounting of the crimes, or punishment of those perpetrators who can be identified, more important if one is attainable only at the expense of the other? If the threat of prosecution would hinder a peace settlement or democratic transition, which has priority?177 How should one rank the values of retribution and cultural autonomy when the population involved sees little worth in retribution and has a longstanding commitment to restorative justice mechanisms?178 Or
175. For a book-length treatment of this idea and some practical applications, see ALAN DERSHOWITZ, RIGHTS FROM WRONGS (2004). For a detailed description and critique of cultural relativism and some postmodern variations, see Eric Blumenson, Mapping the Limits of Skepticism in Law and Morals, 74 TEX. L. REV. 523 (1996). 176. ALEX BORAINE, A COUNTRY UNMASKED: SOUTH AFRICA’S TRUTH AND RECONCILIATION COMMISSION 433 (2000). 177. Perhaps this consequentialist dilemma is best viewed as an instance of value pluralism, without a single correct answer. Interestingly, Thomas Nagel treats the deontological-consequentialist dilemma as a pluralist issue, not because there is more than one valid choice, but because, measuring each by the standards of the other, none are, leaving the actor confronting an intractable moral dilemma in which “there is no honorable moral course for man to take, no course free of guilt and responsibility for evil.” Nagel, supra note 137, at 72. 178. For instance, among the Ifugao people of the Philippines, a murderer is given sanctuary at the household of a sacred leader, and a settlement is negotiated between the clans of the victim and offender. R.F. Barton, Procedure Among the Ifugao, in LAW AND WARFARE: STUDIES IN THE ANTHROPOLOGY OF CONFLICT (Paul Bohannon ed., 1967). The First Nations in Manitoba use healing circles for adults to confess to sexual abuse of children, as part of a larger process of shame, community healing, and forgiveness. John Braithwaite, Restorative Justice: Assessing Optimistic and Pessimistic Accounts, 25 CRIME

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

858

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

when the culture’s belief-system contradicts presumptions underpinning retributive justice?179

the

necessary

& JUSTICE 1, 97 (1999). The traditional restorative justice institution known as Sulha is still used in some Palestinian communities in some cases of murder and other serious crimes. See JOHN BRAITHWAITE, RESTORATIVE JUSTICE AND RESPONSIVE REGULATION 4 (Oxford Univ. Press, 2001). In New Zealand, the institution of Whanau, the system of dispute resolution practiced within the Maori community, has now migrated to the juvenile justice system. See generally GABRIELLE M. MAXWELL & ALLISON MORRIS, FAMILY, VICTIMS AND CULTURE: YOUTH JUSTICE IN NEW ZEALAND (1993). John Braithwaite goes so far as to claim that “restorative justice has been the dominant model of criminal justice throughout most of human history for perhaps all of the world’s peoples.” BRAITHWAITE, supra, at 1; see also Charles Villa-Vicencio, Why Perpetrators Should Not Always be Prosecuted: Where the International Criminal Court and Truth Commissions Meet, 49 EMORY L.J. 205, 211 (2000) (stating that truth telling is at the heart of most African traditional justice systems, and citing Babu Ayindo, Retribution or Restoration for Rwanda?, AFRICA NEWS, Jan. 15, 1998, available at http://web.peacelink.it/afrinews/22_issue/p4.html). Recently, restorative traditions have been successfully resurrected to deal with the aftermath of genocide, atrocities and/or inter-ethnic conflict in Rwanda’s gacaca courts (see generally Peter Uvin, The Gacaca Tribunals in Rwanda, in RECONCILIATION AFTER VIOLENT CONFLICT: A HANDBOOK 116 (David Bloomfield et al. eds., 2003), available at http://www.idea.int/publications/reconciliation/upload/reconciliation_full.pdf); South Africa’s Truth and Reconciliation Commission (see infra notes 183–222 and accompanying text); Uganda’s mato oput rituals (see supra notes 27–31 and accompanying text); East Timor (see generally Carsten Stahn, Accomodating Individual Criminal Responsibility and Reconciliation: The UN Truth Commission for East Timor, 95 AM. J. INT’L L. 952 (2001)); Northern Ireland (see Kieren McEvoy & Harry Mika, Restorative Justice and the Critique of Informalism in Northern Ireland, 42 BRIT. J. CRIMINOLOGY 534 (2002)); and Papua New Guinea (see Sinclair Dinnen, Restorative Justice in Papua New Guinea, 25 INT’L J. SOCIOLOGY L. 245, 256–58 (1997)). 179. This is the claim of the Acholi objectors to ICC prosecutions. See supra notes 27– 32 and accompanying text. The complex beliefs that underwrite retributive justice are common to most criminal justice systems, but they are neither universal nor self-evident. For example, the retributive imperative of punishment is suspect or worse in many faiths, senseless according to many utilitarians, and unduly focused on the defendant and the past according to some restorative justice advocates. See, e.g., JAMES GILLIGAN, PREVENTING VIOLENCE: PROSPECTS FOR TOMORROW 126–28 (2001) (describing a move towards restorative and therapeutic goals in some criminal justice systems); Drumbl, supra note 123, at 551 (“Western justice modalities . . . although globalized, are not universal. They are in fact deeply culturally contingent.”); RAMA MANI, BEYOND RETRIBUTION: SEEKING JUSTICE IN THE SHADOWS OF WAR 47–48 n.128 (2002), cited in Drumbl, supra note 123, at 596 (“[I]nternational justice evidences a predominance of Western-generated theories and an absence of non-Western philosophical discourse.”); Paul van Zyl & Mark Freeman, Conference Report, in THE LEGACY OF ABUSE: CONFRONTING THE PAST, FACING THE FUTURE 3, 6 (Alice H. Henkin ed., 2002) (stating that Asians tend to place more emphasis on acknowledgement and compensation than prosecution, which is not universally accepted as the only just response to past abuse). In the United States, where more than two million people are now incarcerated, dissatisfaction with the retributive model is leading some to look elsewhere. See, e.g., Slobogin, supra note 83, at 121–30; Kenworthy Bilz & John M. Darley, What’s Wrong with Harmless Theories of Punishment, 79 CHI. KENT L. REV. 1215, 1246 (2004) (noting the rapid growth of restorative justice mechanisms in the American criminal justice system).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

859

For the chief prosecutor, law provides some answers where moral reasoning may not. As noted, the Rome Statute’s Preamble describes the ICC’s mission as aimed at ending impunity,180 and Article 53 requires the prosecutor to investigate and prosecute a case unless certain conditions obtain.181 Because one justification for declining to pursue a case is that doing so would serve the interests of justice, the Rome Statute should not prevent the prosecutor from considering a broad range of conflicting considerations when their weight is very great. But his legal and institutional responsibilities do place substantial limits on prosecutorial deference to states that bestow impunity. The prosecutor should not allow a nation’s predilection to forget the past to deflect him from the precisely contrary mission decreed in the Rome Statute, for example.182 But, there is a narrower, yet fundamental, pluralist concern that should occupy the Office of the Prosecutor: the degree of leeway it should afford states that do choose to confront their past and hold perpetrators accountable, but by means of non-penal methods. These have included truth commissions, traditional confession and reintegration rituals, ineligibility for government employment through lustration laws or de facto purges, reparations, and state apologies, among others. Are any of these methods of imposing accountability reasonable enough to warrant ICC deference, at least when they are deemed necessary to serve other vital national interests? The next section examines one truth commission option in detail and suggests that, when properly structured, they can be. III. THE MORAL ACCEPTABILITY OF CONDITIONAL AMNESTIES

We can now consider the programmatic implications of the three preceding arguments, all of which suggest that the prosecutor should be open to at least certain kinds of non-criminal mechanisms. In the following pages I briefly assess one such model, South Africa’s TRC, with its policy of exchanging amnesty for truth, and suggest that it should be deemed morally acceptable for the
180. Rome Statute, supra note 1, pmbl. 181. Id. art. 53. 182. But see Priscilla B. Hayner, International Guidelines for the Creation and

Operation of Truth Commissions: A Preliminary Proposal, 59 LAW & CONTEMP. PROBS. 173, 174, 176 (1996) (stating that when there is a country-wide consensus, “a policy of reconciliation through silence, through trying to forget the past, should be acceptable and accepted by the international community”). Haynar cites postwar Mozambique as an example. Id. at 176.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

860

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

retributive justice, consequentialist, and pluralist reasons adduced above. This is not to imply that all amnesties coupled with truth commissions have been morally acceptable; far from it. But if our question is whether the ICC should ever make room for a non-penal approach to confronting past atrocities, the answer requires that we examine the best case for doing so—and thus far, that case is found in South Africa’s transitional history. The TRC emerged as a compromise during negotiations that arranged the transition from apartheid to democracy. With the apartheid government’s demand for blanket amnesty and the African National Congress’s demand for prosecution of past atrocities each going nowhere, the negotiators settled on a truth commission empowered to grant amnesty to individual applicants who provided a full account of their crimes during apartheid.183 When granted, these amnesties would extinguish the applicant’s civil and criminal liability for these crimes, but also place the case before a committee charged with deciding on an appropriate form of reparation for the victim.184 Those who did not come forward or receive amnesty would remain subject to normal criminal proceedings. This principle of individualized amnesty in exchange for truth distinguished South Africa from Chile, Haiti, Guatemala, and other countries that had granted blanket amnesties to outgoing regimes in exchange for nothing. The South African commission was also exemplary in that, unlike most truth commissions, it was democratically established, and its hearings were both public and widely broadcast.185 How does the TRC fare in terms of the criteria asserted in the preceding arguments? I shall endeavor to show that it does satisfy those criteria. This is a very limited inquiry that must take some central features and aspirations of the TRC off the table: an assessment of the TRC’s efficacy in promoting reconciliation and social repair, providing reparations, uncovering an historical record, promoting democracy and the rule of law, and much more. All of
183. Amnesties were to be granted only if the applicant provided a full disclosure of his crimes, and only for acts that were “associated with political objectives and committed in the course of the conflicts of the past.” The Constitution of the Republic of South Africa, Act 200 of 1993, post-amble; Promotion of the National Unity and Reconciliation Act, Law No. 34 of July 26, 1995, §§ 20(1)(b)–(c) (requiring political objective and full disclosure to be eligible for amnesty) [hereinafter TRC Act]. 184. TRC Act, supra note 183, §§ 20(7)(a), 22(1), 26. 185. No Latin American truth commission held public hearings, and the Guatamalan and El Savadorian commissions were explicitly barred from doing so, allegedly to safeguard sources and witnesses. PRISCILLA HAYNER, UNSPEAKABLE TRUTHS: CONFRONTING STATE TERROR AND ATROCITY 225–26 (2001). Hayner reports that the only commissions that held public hearings as of her writing were those established in South Africa, Uganda, Sri Lanka, and Germany. Id.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

861

these could potentially be relevant to the calculation necessary in a consequentialist or threshold deontological analysis, but there is scarce room for that accounting here. We can only explore one aspect of the TRC’s performance—whether it served as an instrument of justice and accountability for the extreme crimes within its mandate.186 A. Vindicating the Victims

If justice is owed to the victims of heinous crimes, as I have argued above, did the TRC serve to facilitate or hinder South Africa in fulfilling this obligation? Seven in ten black South Africans, the primary victims of the apartheid government, supported its policy of amnesty in exchange for truth, but there were many in the minority, both victims and others, who passionately opposed it.187 The family members of some assassinated activists litigated unsuccessfully to prevent what they saw as a denial of their rights to justice.188 Nevertheless, there is a strong case to be made on behalf of the TRC as an instrument of justice.189 It was surely an imperfect instrument,
186. Assessing the relative merits of criminal and non-criminal modes of accountability raises many empirical issues which cannot be adequately addressed, let alone resolved, here. Many of the studies to date are essentially works of political theory that will be overtaken by events, given the proliferation of these institutions, their variety, and the diverse contexts in which they operate. The arguments here, best considered as hypotheses, are properly elements of a research program to be investigated and revised over time. For information regarding truth commissions and/or the South African TRC in particular, a very partial list of sources includes JAMES L. GIBSON, OVERCOMING APARTHEID: CAN TRUTH RECONCILE A DIVIDED NATION? 4 (2004); HAYNER, supra note 185; TRUTH V. JUSTICE, supra note 40; Ocampo-Minow interview, supra note 23; Llewellyn, supra note 44; Jeremy Sarkin, The Trials and Tribulations of South Africa’s Truth and Reconciliation Commission, 12 S. AFR. J. HUM. RTS. 617 (1996). 187. James L. Gibson, Truth, Justice, and Reconciliation: Judging the Fairness of Amnesty in South Africa, 46 AM. J. POL. SCI. 540, 541, 545 (2002). Others have criticized the therapeutic emphasis of truth commissions. See, e.g., Elizabeth Kiss, Moral Ambition Within and Beyond Political Constraints: Reflections on Restorative Justice, in TRUTH V. JUSTICE, supra note 40, at 68, 72 (arguing that they do “a disservice to survivors who regard themselves not as patients in need of healing but as citizens entitled to justice”). 188. The families of Steven Biko and Griffiths Mxenge, two prominent South African civil rights leaders who were murdered, unsuccessfully challenged the constitutionality of the amnesty provision of the TRC Act in the case of Azanian Peoples Organization (AZAPO) and Others v. President of RSA and Others 1996 (8) BCLR 1015 (CC) (S. Afr.). 189. For different arguments in favor of the TRC as an instrument of justice, see generally Jennifer J. Llewellyn & Robert Howse, Institutions for Restorative Justice: The South African Truth and Reconciliation Commission, 49 U. TORONTO L.J. 355 (1999); JENNIFER J. LLEWELLYN & ROBERT HOWSE, RESTORATIVE JUSTICE: A CONCEPTUAL FRAMEWORK (1998); Gibson, supra note 187; TRUTH V. JUSTICE, supra note 40.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

862

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

but relying solely on criminal investigation and prosecution would have been imperfect as well, in different ways. Justice requires that a state express solidarity with victims of crime, not disregard for them. To do so it must at least investigate, acknowledge, and repudiate the victimization they endured. Criminal punishment generally accomplishes these imperatives—when it can be imposed. The criminal process cannot achieve these goals, however, if the perpetrators are untraceable, or evidence is lacking, or the mass of cases overwhelms the system, and so on. Leaving aside the problem that prosecutions of the old regime may never have been an available option, the case for the TRC rests largely on its comparative advantages over the criminal system in solving these elemental problems. Compared to an exclusively criminal process, the TRC was able to (1) reveal the circumstances and perpetrators of many more crimes; (2) obtain recognition among all races that these events happened and that they constituted criminal atrocities against an oppressed people; and (3) condemn these crimes and censure those responsible, including institutional actors who instigated or were complicit in the atrocities. The first point is crucial, because the essential prerequisite for criminal justice is truth—the truth of what happened, why it happened, and who was responsible.190 On this element of justice, the TRC was clearly more capable, and proved more successful, than the criminal process alone. Trials may generate more credible findings of guilt, largely because of the due process that is applied in adversarial trial systems—the rights to counsel, to present evidence, to cross-examine witnesses, to a high standard of proof, and so on191—but the TRC exposed more crimes, identified more perpetrators, and revealed more truth than would have been
190. On the right to truth, and its importance to many other rights, see Jose Zalaquett, Balancing Ethical Imperatives and Political Constraints: The Dilemma of New Democracies Confronting Past Human Rights Violations, 43 HASTINGS L.J. 1425, 1433 (1992); Comm’n on Human Rights, Sub-Commission on Prevention of Discrimination and Protection of Minorities, Question of the Impunity of Perpetrators of Human Rights Violations (Civil and Political), U.N. Doc. E/CN.4/Sub.2/1997/20 (June 26, 1997) (final report prepared by Mr. Joinet pursuant to Sub-Commission decision 1996/119) (manuscript at 5, para. 17, on file with author) [hereinafter Joinet Report]; Chanfeau Orayce v. Chile, Cases 11.505, Inter-Am. C.H.R., Report No. 512, OEA/ser.L/V/II.98, doc. 7 rev., ¶¶ 85–97 (1997). 191. Diba Majzub, Peace or Justice? Amesties and the International Criminal Court, 3 MELB. J. INT’L L. 251 (2002); Orentlicher, supra note 79, at 2545 n.32; Report to the President from Justice Robert H. Jackson, Chief of Counsel for the United States in the Prosecution of Axis War Criminals, June 7, 1945, reprinted in 39 AM. J. INT’L L. 178, 184 (Supp. 1945) (in which the U.S. chief counsel at the Nuremberg Tribunal describes those trials as a way to “establish incredible events by credible evidence”).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

863

discovered by means of criminal investigations and prosecutions.192 The major reason, of course, is that the offer of amnesty in exchange for truth encouraged thousands of individuals complicit in gross human rights violations to come forward with evidence. Seven thousand people applied for amnesty, many of them obliged to answer questions put to them by victims and their families in public hearings.193 Applicants identified others involved, including superiors who gave them orders.194 The seven to one rejection rate indicates the scrutiny applied to applications.195 Of course, amnesty was a powerful incentive only because a criminal prosecution potentially awaited those who did not reveal the truth. But the conditional amnesty policy revealed the perpetrators of many more crimes, publicized them, and established them as part of a pattern of overwhelming oppression. The smaller reach of an exclusively criminal process would have been compounded by the inevitable plea bargains that generally elicit no testimony, inflict reduced sentences, and receive little public notice. There were additional features of the TRC that made it more effective than criminal prosecutions in uncovering the egregious truths of the past. One is that this was its very purpose: The TRC was designed to establish a complete picture of “the causes, nature and extent of the gross violations of human rights,” along with their “antecedents, circumstances, factors and context.”196 In addition to amnesty applications, the TRC drew on 22,000 submissions from
192. The Human Rights Committee was required to publish “the full names of any person to whom amnesty has been granted, together with sufficient information to identify the act, omission or offence in respect of which amnesty has been granted.” TRC Act, supra note 183, § 20(6). By contrast, Priscilla Hayner observes that “a quick glance at recent transitions . . . makes clear that successful trials of perpetrators are uncommon, and where they do take place, they are usually few in number and rarely reach the most senior perpetrators.” HAYNER, supra note 185, at 89; see also Landsman, supra note 50, at 88–89. For a discussion of the fact-finding defects of trials generally, see Susan Haack, Epistemology Legalized: Or, Truth, Justice, and the American Way, 49 AM. J. JURIS. 43 (2004). 193. Robert I. Rotberg, Truth Commissions and the Provision of Truth, Justice, and Reconciliation, in TRUTH V. JUSTICE, supra note 40, at 3, 15. 194. See, e.g., MINOW, supra note 84, at 59. 195. As of November 2000 (six months before the Amnesty Committee was dissolved), the committee had granted 849 amnesties and rejected 5392 amnesty applicants. Gibson, supra note 187, at 541; see also Dep’t of Justice and Constitutional Development, Amnesty Hearings & Decisions, http://www.doj.gov.za/trc/amntrans/index.htm (last visited June 9, 2005) (providing detailed amnesty statistics and transcripts of the amnesty hearings and decisions). 196. Promotion of National Unity and Reconciliation Act, The Republic of South Africa, Act No. 34 of 1995, Act 95–34, July 26, 1995, as amended by The Promotion of National Unity and Reconciliation Amendment Act No. 84 of 1997.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

864

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

victims and their families, subpoenaed witnesses, and in its later stages, held hearings specifically designed to explore indirect, shared, and institutional responsibility for apartheid.197 Its final report described and condemned the complicity of businesses, media, churches, doctors who routinely misrepresented forensic evidence, and judges who turned a blind eye to deaths in police custody.198 Some have criticized the TRC’s investigation and report as seriously flawed by its many omissions,199 but virtually none of this knowledge would have emerged with any clarity in the criminal process. The purpose of a criminal trial is to determine the guilt or innocence of certain individuals and what truth emerges will be molded by that goal. A trial not only focuses on individuals rather than institutions that may have significant responsibility; it pronounces the latter virtually irrelevant. Most criminal law, including the law that governs the international tribunals, focuses on mens rea, presuming choice and free will.200 It omits the societal input in the crime because of that presumption, so that conditions that encourage or maintain criminal proclivities are nowhere in the picture.201 This is especially problematic if institutions are
197. See Llewellyn, supra note 44, at 98 (regarding hearings on institutional responsibilities); MINOW, supra note 84, at 52–79 (describing the procedures of the TRC). 198. RICHARD WILSON, THE POLITICS OF TRUTH AND RECONCILIATION IN SOUTH AFRICA 35 (2001); Amy Gutmann & Dennis Thompson, The Moral Foundations of Truth Commissions, in TRUTH V. JUSTICE, supra note 40, at 22, 30. These findings are contained in TRC Report, vol. 4. For a discussion of the TRC’s institutional hearings, see ALEX BORAINE, A COUNTRY UNMASKED: INSIDE SOUTH AFRICA’S TRUTH AND RECONCILIATION COMMISSION 145–87 (2000). 199. Mahmood Mamdani says that the TRC failed to “see apartheid as a system, not dedicated towards gross human rights violations per se, but towards dispossession of Africans.” Reconciliation Without Justice, 46 S. AFR. REV OF BOOKS 3–5 (1996). Forced removals, restriction of movement, compulsory labor, and discriminatory regulations that comprised day-to-day life under apartheid received scant attention at the early TRC hearings. See ANTHEA JEFFERY, THE TRUTH ABOUT THE TRUTH COMMISSION 118 (1999) (“In order to accord with established legal principle, it was vital that the TRC take adequate account of all relevant information.”); Kader Asmal, et al., When The Assassin Cries Foul, in LOOKING BACK, REACHING FORWARD: REFLECTIONS ON THE TRUTH AND RECONCILIATION COMMISSION OF SOUTH AFRICA 86–98 (2000); HAYNER, supra note 185, at 100; Francois du Bois, “Nothing but the Truth”: The South African Alternative to Corrective Justice in Transitions to Democracy, in LETHE’S LAW: JUSTICE, LAW AND ETHICS IN RECONCILIATION 91, 108–09 (Emilios Christodoulidis & Scott Veitch eds., 2001). 200. For example, American criminal law requires only “willed” conduct, permitting no inquiry into how “freely willed” the conduct was except insofar as an insanity or duress defense might apply. See generally Powell v. Texas, 392 U.S. 514 (1968) (addressing willfulness for the Eighth Amendment); MODEL PENAL CODE § 2.01 (1962) (defining the act requirement). 201. See JUDITH N. SHKLAR, LEGALISM 172 (1964) (“[T]here is often no mens rea to be found in the development of socially complex events such as war.”); Marvin F. Frankel, The Search for Truth: An Umpireal View, 123 U. PA. L. REV. 1031, 1036 (1975) (“[O]thers

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

865

substantially at fault, or if conduct is widespread rather than deviant, as is often true in the case of crimes against humanity and genocide.202 Truth is the prerequisite for retributive justice, but public condemnation of the crimes, and reaffirmation of the victims’ rights, must follow. The TRC accomplished this in two ways, both realistically unavailable via a criminal trial process. One was a further result of the sheer volume of appalling confessions made to the TRC: As person after person described his own horrific acts, it became virtually impossible for South Africans of all races to deny that these heinous crimes had been perpetrated. With massive numbers of confessions, as opposed to a limited number of contested trials, the TRC was able to establish an authoritative history of deliberate oppression, mutually acknowledged by both perpetrators and victims, where there would otherwise have been different and self-justifying histories prevailing in each racial group. The second advantage the TRC had over a purely criminal trial process concerns the degree of respect and reaffirmation it was able to provide to victims in its procedures. Because the purpose of a criminal trial is to determine the guilt or innocence of the accused, the victim is a means to that end, and not the primary object of interest. The victim’s role in a criminal trial is often confined to testifying, and that testimony is itself restricted by the specific charges that have been brought, the rules of evidence, and by constant interruption. The narrow utilitarian function of the victim’s role may deprive her of an opportunity to recount many of the abuses and the suffering they caused. By contrast, one central, legislativelymandated function of the TRC was to restore “the human and civil dignity of victims of gross human rights violations through testimony and recommendations to the President concerning reparations for victims.”203 Its method was to provide a platform for thousands of South Africans to convey the atrocities they suffered in the nonadversarial environment of the Human Rights Committee (atrocities
searching after facts—in history, geography, medicine, whatever—do not emulate our adversarial system.”); IAN BURUMA, THE WAGES OF GUILT: MEMORIES OF WAR IN GERMANY AND JAPAN 168 (2002) (“[T]rials are not up to dealing with collective responsibility and truth” because the language is “simply wrong.”); Howland & Calathes, supra note 69, at 152; Mark A. Drumbl, Pluralizing International Criminal Justice, 103 MICH. L. REV. 1295, 1309–11 (2005). 202. See generally Drumbl, supra note 84; see also Drumbl, supra note 123, at 571–77, 587–88 (citing Laurel E. Fletcher & Harvey M. Weinstein, Violence and Social Repair: Rethinking the Contribution of Justice to Reconciliation, 24 HUM. RTS. Q. 573, 580 (2002) (“[I]ndividualized guilt may contribute to a myth of collective innocence.”)). 203. TRC Final Report, supra note 125, vol. 1, ch. 4, ¶ 32.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

866

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

that were then made part of the historical record compiled in the volumes of the TRC Report).204 Amnesty applicants not only confessed to their crimes but were also required to answer questions asked by their victims.205 Through their crimes, the perpetrators had asserted a right to mastery over their victims; the TRC’s procedures themselves compelled even the criminals to recognize and repudiate this claim, and affirmed the moral worth and societal standing of the victims.206 This process exemplified what South African political scientist Andre du Toit calls the justice of recognition—“the justice involved in the respect for other persons as equal sources of truth and bearers of rights.”207 According to Pumla Gobodo-Madikizela, a psychologist and member of the Human Rights Committee, many of the victims who testified saw justice in the state’s simple acknowledgment that “you are right, you were damaged, and it was wrong.”208 Reparations by the state were also a means to recognition and reaffirmation of victims, although disappointing to many as executed.209 For all these reasons, the TRC was able to condemn past atrocities and identify their perpetrators on a massive scale—both officially in the TRC reports, and, to an extent that could never have been predicted, in the minds of South Africans generally. Of course, any assessment of the TRC’s record as an instrument of justice must consider the other side of the ledger as well—what it lacked in comparison to a criminal trial system. Two issues are most obvious and troubling. First, the TRC did not provide most of the due process protections afforded to indicted defendants in legitimate criminal
Id. vols. 1–5. Kiss, supra note 187 , at 76. See, e.g., Hampton, supra note 90, at 1684, 1692. Andre du Toit, The Moral Foundations of the South African TRC: Truth as Acknowledgement and Justice as Recognition, in TRUTH V. JUSTICE, supra note 40, at 122, 136. 208. See MINOW, supra note 84, at 60. Darryl Robinson reports that at the Rome Conference, some delegates “were also mindful of the powerful images of sincere repentance by offenders and spontaneous forgiveness from victims in some of the South African TRC proceedings, and wondered whether it would be hubris to declare that the formality of prosecution is the only response in all cases of international crimes.” Robinson, supra note 35, at 483 n.8. 209. In the final volume of the Report, the TRC recommended $270 million in individual compensation for the approximately 19,000 families of victims who testified before the Commission. Following several years of pressure from Archbishop Tutu and the TRC Reparation Committee, President Mbeki announced in April 2003 that the government would provide $85 million in apartheid reparations. Joe Knowles, Government to Pay Families of Apartheid Victims, CHI. TRIB., Apr. 16, 2003, at 10; see also David Crocker, Truth Commissions, Transitional Justice, and Civil Society, in TRUTH V. JUSTICE, supra note 40, at 99, 106.
204. 205. 206. 207.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

867

adjudications, which are designed to insure both justice to the defendant and more highly credible individual verdicts. (Non-penal mechanisms such as gacaca courts in Rwanda and mato oput in Uganda have been heavily criticized on the same ground.210) To some degree, the ICC is in a position to evaluate and enunciate appropriate due process standards for truth commissions, because it can short circuit a state’s conditional amnesty program by bringing its own prosecutions when it finds them lacking. The second issue, of course, is that criminals who confessed were never given a formal punishment, let alone incarcerated. The degree to which confession and the disgrace incident to it are an acceptable alternative form of accountability is the subject of the next section. No mechanism will ever deliver perfect justice. Certainly neither a conventional criminal adjudicatory process, nor the TRC’s amnesty-for-truth program, could have claimed the ability to do so in South Africa. If we ask whether a TRC-type institution should be deemed morally acceptable by the ICC, we must compare its strengths and weaknesses not only with a trial process but also with the goals the ICC can realistically hope to accomplish. One rendition of these goals is offered by the political scientist Catherine Lu: In the end, the ICC’s chief value to the project of moral regeneration may not lie in its punitive powers,
210. Regarding Uganda’s mato oput system, see supra Part I. In Rwanda, the gacaca courts arose in response to the inability of both the decimated domestic judiciary and the ICTR to handle the enormous numbers of post-genocide defendants awaiting trial. Organic Law No. 40/2000, the Gacaca Law, converted Rwanda’s traditional village dispute resolution process into a form of community plea-bargaining that is expected to handle 500,000 to 600,000 criminal cases. Rwandan Justice Officials Predict Sharp Rise in Genocide Trials, BBC MONITORING INT’L REPS., Sept. 11, 2004. In gacaca proceedings, “The entire community will be present and act as a ‘general assembly,’ discussing the alleged act or acts, providing testimony and counter-testimony, argument and counterargument.” Peter Uvin, The Gacaca Tribunals in Rwanda, in RECONCILIATION AFTER VIOLENT CONFLICT: A HANDBOOK 117 (Int’l Institute for Democracy and Electoral Assistance, Apr. 2003). Those individuals who come forward to make an admission of guilt and ask for forgiveness are granted wide reductions in sentencing, and have begun to generate “an avalanche of confessions.” Id. at 117–18. The gacaca system may be the best available solution to the humanitarian crisis taking place in Rwanda’s jails, but it also lacks many of the most basic features thought necessary to safeguard the justice of criminal adjudication. For example, defendants are not provided counsel; the courts have no presumption of innocence, rules of evidence, or rules of procedure; and many judges lack necessary training, and are directly related to a victim or perpetrator of the genocide. See Amnesty International, Gacaca: A Question of Justice (Dec. 2002), available at http://web.amnesty.org/library/index/engafr470072002; Leach Werchick, Prospects for Justice in Rwanda’s Citizen Tribunals, 8(3) HUM. RTS. BRIEF 15, 16–17 (2001). Werchick also reports that “unchecked popular participation in the prosecution and judging of defendants jeopardizes key guarantees of judicial independence and impartiality.” Id. at 28.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

868

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

but in its ability to establish an authoritative and truthful record of events and violations, to promote reparations for victims and the affected communities, and to encourage the public and private reflections required for moral regeneration in all survivors of violence and atrocity.211 By this standard, the TRC and its successors, which are especially suited to these goals, offer one route towards fulfilling the ICC’s potential. B. Holding Perpetrators Accountable

As described in Part II.B.1, retributive justice can be conceived in alternative ways. We have just considered one conception, which focuses on the imperative of solidarity with the victims of crime. Other retributivists define the ultimate requirement of justice differently, as an obligation to bring perpetrators to justice.212 But even on this conception, there are reasons for the ICC, as a global institution, to allow states some leeway in the approaches it invokes to do so. Clearly the TRC did not impose criminal punishment on successful amnesty applicants. But if the ICC is to embrace an appropriate degree of pluralism, it should consider whether the TRC, with its amnesties conditioned on truth, imposed an alternative but reasonable form of accountability that should be acceptable in future cases. Again, the most powerful argument in favor of the TRC as a mechanism of accountability starts with its greater reach—its ability to identify far more perpetrators than a reliance on investigations and prosecutions alone would have. The result was to fix responsibility for vast numbers of crimes, and also impart a measure of accountability on those who committed them, in two different ways. For those who did not come forward but were named by others who did, there could be criminal prosecution and punishment with evidence that would not have existed otherwise. For those who came forward and received amnesty in exchange for their confessions, there was a different kind of accountability: disgrace. TRC hearings were public—covered in great detail by the print media, and broadcast on television and radio for hours a day. A one-hour
211. Catherine Lu, The International Criminal Court as an Institution of Moral Generation: Problems and Prospects 30 (Dec. 7, 2004) (unpublished manuscript, on file with author). 212. See supra notes 92, 107–12 and accompanying text.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

869

weekly television show on the week’s hearings garnered more viewers than any other current affairs program.213 The visibility of these hearings made them a powerful form of public disgrace for amnesty applicants that in important ways mimicked the exposure of a criminal trial. Admissions often came from respected figures, including doctors and scientists as well as leaders in military and law enforcement.214 The shame that went with this public exposure of heinous criminal acts assured that, like criminal convictions, those found responsible suffered a strong form of moral condemnation, including stigmatization and censure from the communities to which they returned.215 At the least, disgrace is not impunity. It is one means of holding an offender accountable. Justice Richard Goldstone, formerly a South African Justice and Chief Prosecutor of the ICTY, believes that amnesty applicants “suffered a very real punishment” by the shame they suffered following their public confessions.216 Many traditional, honor-based societies continue to exist, and for their members shame is an extremely powerful form of punishment. Even in a modern society like the United States, there is growing attention to non-incarcerative shaming punishments, which some scholars argue can serve all of the purposes of punishment, including retributive justice,217 and others criticize as too harsh.218 Of course,
213. HAYNER, supra note 185, at 226. 214. Crocker, supra note 209, at 104. 215. See, e.g., Archbishop Desmond Tutu, Criticisms and Challenges, in 1 TRUTH AND

RECONCILIATION COMMISSION OF SOUTH AFRICA, FINAL REPORT ¶ 35 (1998). Argentina’s Truth Commission report similarly changed the lives of many of those whose responsibility was revealed. See Crocker, supra note 209, at 104. 216. Gibson, supra note 187, at 544; see also, Crocker, supra note 209 at 102–03 (arguing that truth commissions can impose accountability through the sanction of shame). 217. See, e.g., Dan M. Kahan, What Do Alternative Sanctions Mean?, 63 U. CHI. L. REV. 591, 630–53 (1996); Hampton, supra note 90, at 1695 (stating that through shame “we are humbled . . . [and] chastened, just as surely as if we had been punished”); Toni M. Massaro, Shame, Culture and American Criminal Law, 89 MICH. L. REV. 1880, 1890 (1991) (but criticizing shame punishments on other grounds). For examinations of shaming punishments generally, see generally JOHN BRAITHWAITE, CRIME, SHAME AND REINTEGRATION (1989); Stephen P. Garvey, Can Shaming Punishments Educate?, 65 U. CHI. L. REV. 733 (1998); Dan Markel, Are Shaming Punishments Beautifully Retributive? Retributivism and the Implications for the Alternative Sanctions Debate, 54 VAND. L. REV. 2157 (2001). 218. There are some well-founded objections to punishments designed to shame the offender, which can range from orders to wear an “I am a drunk driver” sign in public to public floggings in Iran. Manuchehr Sanadjian, A Public Flogging in South-Western Iran: Juridical Rule, Abolition of Legality and Local Resistance, in I NSIDE AND O UTSIDE THE LAW 157, 157–68 (Olivia Harris ed., 1996). One objection is that these techniques involve purposeful cruelty and humiliation, which is itself a violation of human dignity. See, e.g., Massaro, supra note 217, at 1942–43. Another, advanced by James Whitman, is that

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

870

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

these are formal punishments imposed by a court after conviction, while the shame attendant on confession before the TRC was not. But shame does serve as one form of accountability in either case, and a form that is particularly geared to the degree of the perpetrator’s wrongdoing. Amnesty in exchange for public confession imposes accountability in another way as well—internally, through recognition of one’s own guilt. Antony Duff argues that this should be understood as “a punitive process, which seeks to impose suffering on offenders for their offences, or more precisely to persuade them to impose such suffering on themselves.”219 Adjudications of guilt imposed upon a defendant may not shake his/her self-righteousness or resentment, whereas a public confession before one’s own family, friends, and community engenders awareness of one’s extreme transgression, and often remorse.220 Some Ugandans describe the Acholi traditional mato oput ritual as a form of punishment in this way, because “you are free, but feel the weight of what you’ve done.”221 Many may doubt that this type of punishment is enough, at least when torture, assassinations, or crimes against humanity are involved. Professors Gutmann and Thompson argue that justice is
shaming-punishments encourage uncontrollable, vigilante-type responses from the public. James Q. Whitman, What is Wrong with Inflicting Shame Sanctions?, 107 YALE L.J. 1055, 1088–89 (1998). But these objections can have no bearing on public exposure of one’s crimes; shame is certainly involved, but not in an objectionable way. The purpose is not humiliation but rather public recognition and condemnation of extraordinary victimizations. No measures beyond this were inflicted in South Africa, and to say that mere publication is objectionable is hardly distinguishable from a policy of mandatory, official secrecy regarding extreme crimes. Moreover, what shame accrued to the offenders was similar in kind to the shame that generally accompanies public notice of a criminal conviction. See, e.g., James Q. Whitman, Making Happy Punishers, 118 HARV. L. REV. 2698, 2710 (2005) (“Even the most dedicated critics of shaming . . . must acknowledge that there will always be some unavoidable element of shame and stigmatization in the law, especially in the criminal law”); Duff, supra note 95 (Punishment is designed to “communicate to offenders the censure or condemnation that they deserve for their crimes.”). 219. Antony Duff, Truth, Reconciliation and Punishment 4 (manuscript on file with author). According to Duff, the suffering imposed “includes the suffering involved in being confronted with and recognizing one’s wrongdoing, the remorse which that recognition must bring with it, and the burdensome reparation through which apology is to be expressed: this is suffering that offenders deserve to undergo, and that the criminal mediation process aims to induce.” Id. 220. See Drumbl, supra note 84, at 1265. As Drumbl notes, “When aggressors can see the hurt for themselves instead of denying it in the splendid insularity of prison, can hear the words of survivors, and can look at mass graves instead of jail walls, perhaps then their consciences will become troubled.” Id. at 1262. 221. Refugee Law Project Paper 15, supra note 25, at 26–27 (quoting an Acholi leader).

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

871

not achieved when a murderer publicly acknowledges his crimes and is then allowed to walk the streets without trial or formal punishment.222 The argument here is one level removed from their conclusion; it is that the exposure and shame imposed on each applicant through the TRC’s amnesty process is a genuine form of accountability, and a genuine departure from impunity, and as such, within reasonable pluralist limits—within the margin of appreciation, so to speak—that should guide the ICC as a global institution. CONCLUSION I have endeavored to inventory a number of difficult issues that underlie what is typically referred to as the “peace versus justice” debate, and to develop an analysis that can help resolve them. At the least, I hope that I have shown that an abiding concern for human rights will sometimes warrant looking beyond prosecution to alternative possibilities. Notwithstanding the achievements and great potential of the ICC and the other international criminal tribunals, one cannot assume that uniform reliance on prosecution and punishment will always best further the cause of human rights. Such institutions as the South African TRC, with its amnesties conditioned on confession, should also be recognized as a human rights advance, and in certain circumstances a necessary and morally acceptable option for the ICC. As this Article demonstrates, three reasons support this view. The first concerns the nature of the ICC’s obligation to do justice. I have argued that there is such an obligation because it is not morally tolerable for a society to ignore heinous crimes against its members. A corollary is that this obligation is fundamentally a duty to repudiate the crime and reaffirm the inviolable moral status and equality of the victims. Punishment of the perpetrator is one way of achieving this, but such alternative methods as the South African Truth and Reconciliation Commission and some traditional justice systems have done so as well. Consequently, in particular circumstances the ICC would be warranted in deferring to states that choose to invoke certain non-penal approaches to condemning crimes and imposing accountability. The second reason emerges when prosecution threatens to inflict grave injury to innocent third parties. This is the contested claim of many in the victim community in Northern Uganda—that
V. JUSTICE,

222. Gutmann & Thompson, The Moral Foundations of Truth Commissions, in TRUTH

supra note 40, at 22, 25.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

872

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

ICC charges against Kony and others who assuredly deserve prosecution would also destroy the prospects of a peace settlement, and keep the death squads in business. I have considered several approaches to addressing conflicts between penal justice and peace or other important stakes, and suggested that even if we assume that justice requires a prosecution, in grave cases this duty to prosecute can be outweighed and must give way. But if it does give way, there remains a duty to mitigate the resulting injustice to the fullest extent possible, and here again, truth commissions and conditional amnesties can play an important role. The final reason the ICC should not foreclose all non-penal approaches lies in the wisdom of tolerance amidst global diversity. As described in Part II.C, there are procedural, substantive, and pragmatic reasons why the ICC would do well to adopt a pluralist philosophy in its interests-of-justice and complementarity assessments. One of the central challenges in developing a global standard of justice, as the ICC will inevitably do through its complementarity assessments, is defining the proper scope and necessary limits of diversity in state approaches to accountability. No one has been able to precisely locate the points at which legitimate moral diversity ends and universal moral imperatives begin, but it is a central issue of our time, and an unavoidable one for the Court.223

223. To return to the decision matrix with which we started, the three arguments above, coupled with the obligation to mitigate discussed supra at notes 36–53 and accompanying text, suggest the following responses to each of the six scenarios:

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

2006] THE ICC: CHALLENGE OF A GLOBAL STANDARD OF JUSTICE

873

As I have noted, our subject is often characterized as the “peace versus justice” question. This terminology is unfortunate because it portrays the issue as a moral dilemma with no solution, and confines one’s thinking to a forced choice between war and injustice. Part of the burden of this Article has been to show that even in the most difficult situations, there can be a third alternative beyond war or injustice, if the ICC embraces it. Some may fear that unless the Court consistently demands criminal prosecution, its potential contributions to international justice will be thwarted. I have suggested instead that the ICC must sometimes play a role in ensuring that there are less abysmal alternatives than peace without

Of course, this framework only begins an analysis that may properly end up suggesting different choices. One obvious reason is that our inquiry has been confined to certain significant and recurrent moral factors, but in any particular case, the decision whether to prosecute will simultaneously serve or disserve numerous additional interests as well— prosecutorial even-handedness, maintaining the credibility and legitimacy of the institution and international criminal law, husbanding resources, and so on—which must be put in the balance. Most notably, this inquiry has bracketed the utilitarian rationale for the ICC, resting on its potential as an instrument of deterrence. A second reason that moral conclusions alone are insufficiently determinate in practice is that there may be instrumental questions about which course of action is the best means of serving them. Moreover, if “ought” implies “can,” moral conclusions themselves may need to be modified by the specific constraints and opportunities confronting the Court at the operative time. Finally, of course, the prosecutor also will have to attend to certain legal controversies concerning the scope of his discretion in light of the Rome Statute and certain human rights treaties.

BLUMENSON ARTICLE FINAL2.DOC

5/9/2006 3:35 PM

874

COLUMBIA JOURNAL OF TRANSNATIONAL LAW

[44:801

justice or justice without peace, and that doing so will further rather than hinder one of its essential obligations: To insure that those who suffer extreme crime will no longer have to face an international community that is indifferent to their victimization. The history of human rights is the story of an inherently slow journey from relations of power to relations of justice. It has to be, because power does not give up its prerogatives easily or unconditionally. For that reason, the ICC cannot expect to escape severe constraints and complex moral dilemmas, but it can expect to be an integral part of the movement towards international justice.

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close